(work on title)Visual learning and the necessary nap
Naps prevent perceptual deterioration and facilitate learning in local
networks of human visual cortex
A thesis presented
By
Sara Carole Mednick
To
The Department of Psychology
In partial fulfillment of the requirements
For the degree of
Doctor of Philosophy
In the subject of Cognition, Brain, Behavior
Harvard University
Cambridge, Massachusetts
May 23, 2003
INTRODUCTION
Humans and other animals are constantly and rapidly learning from the environment that surrounds them. We learn spatial layouts of rooms and neighborhoods, auditory patterns in music and language, detailed configurations of faces, and much more, all without conscious effort or much apparent attention. We learn an extraordinary amount of information, but not all information is learned in the same way, or by the same mechanisms, or brain areas.
One of the main goals of cognitive neuroscience is to understand how we learn. To this end, researchers have recently investigated the effect of sleep on learning and found that a wide range of learning is dependent on post-training, nocturnal sleep, including visual perceptual, auditory and motor learning. In fact, some studies find that improvement on a visual texture discrimination task occurs only after subjects have slept for at least six hours (Stickgold, 2000a). Findings such as these open the door to new areas of research and invite new questions. One such question is: why do people nap? and do daytime naps have a measurable benefit for learning similar to nocturnal sleep?
In this dissertation, the effect of napping on within-day repeated testing on a texture discrimination task is investigated. I examine whether napping is an effective tool for within-day maintenance of optimal performance on a visual perceptual task; whether naps, similar to nocturnal sleep, can facilitate learning; I compare the benefits of napping to nocturnal sleep. These investigations on napping occur against the backdrop of a second investigation into performance deterioration as a result of repeated, within-day testing on a texture discrimination task.
Overview
The most important new findings in a field can sometimes originate outside the field. These findings usually bring together areas of research not previously considered similar, as well as incorporate techniques not common to the field. Sometimes these new findings can bring about a paradigmatic shift in thinking and scientific approach. Such a shift may be occurring in the field of memory research with the study of the effects of sleep on learning (for review, (Maquet, 2001).
Traditional methods for studying learning have focused on what we learn (faces, spatial layouts, and grammatical rules) in order to understand memory representation, as well as to study who learns what (amnesics, lesion patients, or people with schizophrenia). One goal of these studies is to understand where in the brain such representation is stored. Understanding what and who has been important for laying the ground work for the study of memory and learning, such as the development of the multiple memory system model, which separates declarative from non-declarative memories by their relative dependence on the hippocampal formation. This ground work helps make it possible to begin examining the mechanism of learning and memory.
Recent developments in research on sleep and memory contribute to our understanding of mechanisms. The approach of this research has been to ask not just when, but also under what circumstances does learning occur. Examples of circumstances that influence learning are: circadian rhythm, sleep deprivation and/or amount of sleep. Results from this research have demonstrated that some learning occurs slowly and depends on sleep, while other memories are more rapidly formed and do not rely on sleep. Understanding the contribution of sleep to learning illuminates the biological processes and neurophysiological changes (e.g. protein synthesis, alterations in neurotransmitters and electrical brain waves) that may be important and necessary for learning to occur.
Questions may also now be asked that consider new aspects of the relationship of sleep and learning, such as what happens to performance with and without sleep, the quality and quantity of sleep that is necessary for learning, the type of learning that requires sleep and the type of sleep that contributes to learning. In my dissertation, these are the questions I will be asking.
A brief overview of learning and memory research
Since the time of Plato and Aristotle memory has been a widely studied area of psychology. Experimental studies of memory began in the early 1900s, when Ebbinghaus began his now famous studies of his own memory by tracking how many rehearsals he needed to recall a list of nonsense syllables, methods that are still used in modern memory tests. The birth of the multiple memory system began with Broadbent’s (1958) information processing approach which explicitly contained a limited capacity, short-term memory store where information could either get rehearsed and become part of a long-term memory store or else decay into forgetfulness.
In the mid-fifties doctors removed the hippocampal complex and parts of the temporal lobe bilaterally of an epileptic patient named H.M. and thus accidentally revolutionized the neuroscience of memory research (Scoville, 1957; Milner, 1962). The removal of the hippocampal complex helped H.M.’s epilepsy, but also permanently damaged his memory so that he could not remember new experiences (anterograde amnesia), and he experienced memory loss for events that occurred up to a few years before his surgery. The study of amnesic patients like H.M. revealed a role of the medial temporal lobe (includes the hippocampus, the dentate gyrus and subiculum, as well as complementary areas of the entorhinal cortex, perirhinal cortex and parahippocampal gyrus) in memory acquisition as well as in long term memory consolidation (Weiskrantz and Warrington 1979; Schacter 1985).
Amnesic patients, such as H.M., are living proof of another natural division in memory between explicit and implicit memory. These patients, with severe medial temporal lobe damage, evidence marked deficits in explicit memory tasks that require conscious encoding and retrieval of items, but are unimpaired on implicit memory tests, which access memory for items that have not been consciously encoded and are, therefore, not available for explicit recall. A memory model had been constructed that consisted of a hippocampal-dependent, declarative (explicit) memory system and a non-hippocampal-dependent, non-declarative (implicit) memory system, both of which are supported by a wide range of brain areas (Schacter and Tulving 1994; Squire 1994).
The standard model of declarative memory includes the subtype of episodic memory (comprising knowledge of personal events or episodes), and the subtype of semantic knowledge (comprising knowledge of “facts” about the world: name of the first president, public events and personalities, as well as conceptual knowledge of words, grammars and objects).
All of the other memories that do not require conscious acquisition and recall are part of a collection of non-declarative and procedural memories. These memories includes information acquired during skill learning (including motor skills, perceptual skills, and cognitive skills) and habit formation, simple classical conditioning, as well as priming and non-associative learning. None of these memory categories are thought to depend on the hippocampus, as amnesic patients are unimpaired in these areas. Procedural learning can be found across all sensory modalities, visual (perceptual learning), auditory (music learning and classical conditioning), tactile (pain conditioning), gustatory (taste aversion), and smell (animal studies of maze learning), as well as language (grammar rule learning), and motor learning (sequence learning).
Until recently, the study of memory and learning has been dominated by the methodologies of declarative memory research and the functions of both declarative and non-declarative memory were evaluated similarly. Subjects were exposed to a stimulus once, followed by a test of retention, either overtly or covertly. Or alternatively, a stimulus was repeatedly shown and the number of exposures necessary for consolidation of the stimulus was calculated, either by explicit or implicit response. If learning was achieved, researchers could then examine deterioration of the memory representation from interference or decay.
Recently, a new approach to examining the mechanism of learning has broken the mold of prior memory research. These studies investigate whether other factors may contribute to learning such as slowly developing, off-line processes that produce long-lasting behavioral changes. These studies have examined experience-dependent plasticity of the cortex (Gilbert, 2001), the effect of electrical stimulation of the brain (McGaugh et al, 1979), activation of genes associated with plasticity (Herrera et al, 1996), neural development (Frank et al, 2001), and the effect of sleep on sensory-motor learning (Walker, 2002; Karni, 1991; Stickgold, 2000a). As this dissertation focuses on visual perceptual processing and sleep, the next section will give an overview of visual perceptual learning.
Visual perception and learning
Research in visual perception has formed the basis for much of the work in learning and memory, from early studies of visual short-term memory by Sperling (Sperling 1960) to present work in perceptual learning. Over the past twenty years a special kind of learning has been identified and studied; this learning shows distinct signs of isolated plasticity of early visual areas that may or may not depend on higher areas of processing. This has been termed perceptual learning, as it does not appear to entail higher cognitive processing, but instead shows improved performance on very basic psychophysical measures, such as discrimination of orientation, spatial frequency, motion signal and vernier acuity.
Perceptual learning is characterized as being highly rigid, in that most studies find that improved performance on one task does not generalize to other similar tasks (Gilbert et al. 2001; Ahissar and Hochstein 1997). Learning has been shown to be specific to the trained stimulus (e.g. oriented lines, spatial frequency) (Fiorentini and Berardi 1980; Crist, Kapadia et al. 1997), the retinotopic area of the visual cortex (Fahle, Edelman et al. 1995; Crist, Li et al. 2001), and, in some cases, the trained eye (Fahle, Edelman et al. 1995; Schwartz 2002; Karni et al, 1991). The processes underlying this learning presumably involve mechanisms of experience-dependent cortical plasticity occurring in primary visual cortex (Zohary, Celebrini et al. 1994). Learning has also been shown to generalize to untrained stimuli in circumstances where a wide range of stimuli are trained (Ahissar and Hochstein1997; Liu, 2000). Many models have been proposed to explain the specificity question in perceptual learning. The rich mapping of the visual cortex allows for the marriage of both structure and function in the modeling of perceptual learning.
Wiring of the visual cortex
Orientation selectivity is a remarkable property of neurons in the visual cortex which are thought to provide the detection of local bars and edges in processed visual images and encodes their orientations (Hubel and Wiesel 1974). According to the concept of columnar organization, neighboring neurons in the visual cortex have similar orientation tunings and comprise an orientation column (Hubel and Wiesel 1974). Columns close together generally have similar, but not identical, orientation preferences, and distant columns generally have more dissimilar preferences. For orientation preferences, the arrangement of cells forms an orientation map of the retinal input (Blasdel 1992). Each location on the retina is mapped to a region on the orientation map (i.e., retinotopic specificity), with each possible orientation at that location represented by different orientation selective cells. The global layout of the orientation map (and consequently the orientation preferences of the individual neurons in the map) is formed with experience during early development (Hubel and Wiesel 1968; Blakemore, 1970; Hubel and Wiesel 1974; Blakemore, 1975).
An important addition to the structure and function of the visual cortex is a network of extensive, long-range lateral connections between neurons in neighboring columns with similar preferences (Gilbert et al, 1983; Gilbert et al, 1990). This is a network of connectivity formed by the axons of cortical pyramidal cells. The lateral connectivity is not uniform or genetically determined, but develops based on visual experience (Katz et al, 1992; Burkhalter 1993; Rubin, 1997; Dalva et al, 1994). These connections are initially widespread, but develop into clustered patches at approximately the same time as the orientation maps form. The lateral connections are far more numerous than the afferents and they are believed to have a substantial influence on cortical activity.
An important factor in understanding plasticity underlying perceptual learning in the visual cortex is to understand the temporal characteristics of the learning. Depending on the task, plasticity underlying learning can occur both over relatively short periods of training, as in the “eureka effect” (seconds to minutes) (Ahissar and Hochstein 1993; Fahle, Edelman et al. 1995;Rubin, 1997), as well as over periods of days (Karni and Sagi 1993; Stickgold, Whidbee et al. 2000b). This slow improvement occurs even in the absence of continued practice, indicating that some off-line processing must occur. Though many models have been proposed, the actual mechanisms of visual cortical plasticity involved in both the fast and slow improvement remain almost completely unknown. Studies that investigate not only what is learned, but also examine under which circumstances fast and slow learning occur, have found evidence that some of the slow learning is sleep-dependent (Karni and Sagi 1993; Karni, Tanne et al. 1994; Karni, Weisberg et al. 1995; Gais, Plihal et al. 2000; Stickgold, James et al. 2000a; Stickgold, Whidbee et al. 2000b; Stickgold, Hobson et al. 2001).
Sleep-dependent learning and plasticity
Sleep-dependent learning has been demonstrated across a wide range of sensory and motor tasks (for review see, Maquet 2001), some of the clearest evidence of sleep-dependent learning comes from studies of a texture discrimination task originally developed by Karni and Sagi (1991). Karni and colleagues have shown that improvement is only evident several hours after training, and that improvement can develop overnight, although only when rapid eye movement (REM) sleep is allowed (Karni, Tanne et al. 1994). (A further discussion of sleep architecture can be found in the next subsection.) Karni et al. subsequently showed in an fMRI study that training across days - weeks could lead to enlarged regions of activation in the primary and secondary visual cortex (Karni, Weisberg et al. 1995). This finding was later replicated by Schwartz et al. (2002) over a 12hr period.
In a series of subsequent studies, Stickgold and colleagues demonstrated that improvement in performance on the texture discrimination task can be achieved only after a full night's sleep (2000a,b). Without a post-training night of sleep, no learning occurs even after subjects are allowed two nights of recovery sleep. This consolidation process continues beyond the first post-training night without further training, such that texture discrimination performance tested four days after the initial testing is the same in a group tested once a day for four days as a group tested only once on the first day and once on the fourth day (2000b). By examining the effect of particular sleep stages on learning, Stickgold noted a relationship between overnight improvement and both deep slow wave sleep (SWS) and rapid eye movement (REM) sleep, specifically improvement correlated with the product of the amount of SWS in the early part of the night and REM in the last part of the night (2000a). Stickgold et al proposed a two-step model for sleep-dependent learning, in which SWS and REM have independent and sequential roles in the process of consolidation. As such, it explains why a full night of sleep (6 to 8 hours) is required for optimal consolidation of post-training learning. The next section briefly reviews sleep structure.
The structure of nocturnal sleep.
Sleep is a highly structured set of processes separated into five phases, each demonstrating stereotypic electrical activity, neuro-chemical expressions, and enhancement and depression of varying brain regions. The five phases, stage 1, 2, 3, 4 , and Rapid Eye Movement (REM), progress in a cycle from stage 1 through stage four and then back up to REM sleep (Figure 1). We spend almost 50 percent of our total sleep time in stage 2, about 20 percent in REM sleep, and the remaining 30 percent in the other stages. Infants, by contrast, spend about half of their sleep time in REM sleep. The duration of an entire cycle lasts for 90-110 minutes. The period of deepest sleep, slow wave sleep (SWS), is composed of stages 3 & 4. In stage 3, extremely slow brain waves called delta waves begin to appear, interspersed with smaller, faster waves. By stage 4, the brain produces delta waves almost exclusively. The beginning of the night is characterized by a larger proportion of SWS. As the night progresses the amount of SWS decreases and there is a corresponding increase in REM sleep. Consequently, the morning period is rich in REM sleep. REM sleep, in contrast to SWS, is a lighter sleep accompanied by rapid irregular shallow breathing, rapid jerking eye movements, increases in heart rate, as well as limb muscle paralysis.
Neuromodulator fluctuations occur across different sleep stages. Brainstem systems that control the REM-NREM (non-REM) cycle include the noandrenergic (NE) locus coerlus, the serotonergic (5-HT) dorsal Raphe nucleus, and the cholinergic (ACh) nuclei of the dorsolateral pons (Hobson 1975). Whereas NREM is characterized by decreases in all three neuromodulators compared with waking, ACh levels in REM are equal to or higher (Karnetani, 1990) than during wake, and levels of NE and 5-HT drop to zero. Changes in both ACh and 5-HT have been proposed as mediators of memory consolidation (Hasselmo 1999; Graves, Pack et al. 2001). Table 1 shows physiological correlates of the different sleep stages.
|Table 1: Physiological Correlates of Sleep Stages |REM |Stage 2 |SWS |
| | | | |
|Synchronous brain electrical activity |4 to 6 Hz |12 to 14 Hz |0.5 to 4 Hz |
|Eye movements |++ |-- |-- |
|Muscle tone |-- |- |- |
|Cholinergic Modulation (ACh) |++ |- |- |
|Aminergic modulation (NE & 5-HT) |-- |- |- |
Neurophysiological basis of sleep-dependent learning
An important key to the puzzle of sleep-dependent perceptual learning links development of the visual system with sleep. As a predecessor to such a finding, Hubel and Wiesel discovered that normal brain development was experience-dependent by demonstrating that monocular deprivation during a critical period in kittens prevented normal development of brain circuitry in the visual cortex (Hubel et al, 1978). They showed that the cortical area corresponding to the deprived eye developed much narrower ocular columns compared with the non-deprived eye. In addition, neurons in the visual cortex that usually responded to input from both eyes, no longer respond to inputs from the eye that was deprived.
Recently, Frank, Issa and Stryker contributed a missing piece to the work of Hubel and Wiesel by examining the role of sleep in ocular dominance plasticity (Frank, Issa et al. 2001). The researchers examined how sleep after monocular deprivation affects cortical representation of ocular dominance for both eyes. They found that sleep greatly enhanced cortical plasticity of ocular dominance after monocular deprivation, whereas sleep deprivation completely prevented the enhancement of cortical plasticity that was observed in the sleeping cats. Thus sleep can play a direct role in the development of ocular dominance columns during the critical period in the kitten’s development. Human babies sleep a tremendous amount during development, particularly copious REM is observed in young in both humans and animal (Roffwarg, Muzio et al. 1966). In the adult human, sleep has been shown to be necessary for retinotopically specific improvement on the texture discrimination task (Stickgold, 2000b; Gais, 2000) and for sequence learning in a finger tapping task (Walker et al, 2002). Together these findings demonstrate a mechanism of learning in the brain that deserves considerable investigation.
What about napping?
Most of the studies that have examined the role of sleep on learning in humans, have studied nocturnal sleep. One study reported that at least six hours of sleep are necessary before any learning can occur (on the texture discrimination task) (Stickgold, 2000a). One wonders how these nocturnal sleep findings would hold up in studies of short day time sleep (naps).
Throughout history to the present day, many cultures have traditionally taken a brief “siesta” in the middle of the day. In our accelerated and stressful work-oriented life, the siesta, although still present, has been foreshortened to the “power nap” to accommodate our increased work schedules (Stein, 2001). Such power naps appear to be used by individuals as a way to refresh the mind and recharge the system. But these descriptions are only metaphors to disguise the fact that the benefit of napping is unexplored territory in the learning and sleep literature.
Strategic napping has been shown to improve alertness, productivity, and mood (Dinges and Broughton 1989; Takahashi and Arito 2000). This is especially so under sleep-deprived conditions (Bonnet 1991; Bonnet and Arand 1994), in nightshift work (Rosa 1993), and during prolonged periods of driving (Horne and Reyner 1996). Naps have also been shown to enhance psychomotor speed as well as post-nap, short-term memory acquisition (Taub 1979; Harma, Knauth et al. 1989). All of these examples are, perhaps, unsurprising as performance has been related to alertness, and naps, in most cases, can be expected to decrease sleepiness and increase alertness. Current models of the role of sleep in learning do not address or even allow for the possibility that a brief, mid-day nap may have benefits to human cognition and memory consolidation similar to nocturnal sleep.
Thus, in reaching the goal of understanding the learning process, understanding the role of the nap in sleep-dependent changes in human performance appears a fruitful challenge. This dissertation addresses this issue by studying the effect of daytime napping on repeated, within-day testing on a texture discrimination task. Our studies demonstrate: 1) that human performance on a visual task does in fact deteriorate with repeated testing, 2) this deterioration can be reversed only with sleep, and not quiet rest with eyes closed, increased motivation, or decrease in task difficulty, 3) that the deterioration is experience-dependent and does not dissipate with time, and 4) that a nap is important for maintenance of human performance and this maintenance is driven by SWS. We also show that 5) a nap containing both REM and SWS can facilitate perceptual learning equal to that of nocturnal sleep, and that this nap-dependent learning is additive with a night of sleep.
GENERAL METHODS
The texture discrimination task involves identifying the orientation of an array of three diagonal bars against a background of horizontal bars (Karni et al, 1991) (See Fig. 2). Each trial begins with a fixation point at the center of a computer screen on which subjects keep their eyes focused. Subjects then press the space bar to start the trial and the target screen is then briefly displayed (17ms). The display contains a fixation letter (either an “L” or a “T”) in the center of a field of horizontal dashes, as well as a series of three diagonal dashes located in the bottom left quadrant of the screen that are in a horizontal or in a vertical array. For some studies the peripheral target is located in the bottom right quadrant. After the target screen appears there is a brief blank screen (presented for varying durations across blocks) followed by a masking screen. Subjects make two discriminations: 1) an easy fixation discrimination (was it an L or T?), which ensures fixation; and 2) a peripheral discrimination (was it a horizontal or vertical array). Blocks of 50 trials are carried out with shorter and shorter durations of the inter-stimulus interval (ISI) between presentation of the target screen and the subsequent masking screen. From a series of 1,250 trials, an interpolated target-mask “threshold ISI” is calculated at which a subject can identify the orientation of the array with 80% accuracy. Improvement is defined as the decrease in threshold ISI between training and re-test. Deterioration in performance is defined as an increase in threshold ISI between training and re-test.
Polysomnographic (PSG) sleep recording
Subjects, who participate in a daytime nap experiment, have their sleep recorded polysomnographically (Rechtschaffen and Kales 1968) with standard electroencephalographic (EEG), electro-oculographic (EOG) and electromyographic (EMG) measures. Sleep stages are correlated (Pearson’s regression analyses) with task improvement.
Statistical Methods
Individual response accuracy and reaction times are recorded for each trial. Statistical procedures used for these data consist of standard statistical tests (mixed model ANOVAs and both paired and unpaired t-tests) for the discrimination of differences between conditions and groups and regression analysis for within group analysis.
Overview of protocols
Study 1A examines how performance on the texture discrimination task changes with repeated, within-day testing, and the effect of both a long nap (one hour) and a short nap (half an hour) on performance. Study 1B examines whether sleep is necessary for reversing performance deterioration or whether quiet rest with no visual stimulation may also be effective. In Study 1C, we ask whether the deterioration is due to decreases in motivation of the control subjects by offering monetary reward for improving performance. In study 1D, we attempt to localize the deterioration to early visual cortex by testing for retinotopic specificity. In Study 1E, we test whether the deterioration is due to the task difficulty. We do this by testing subject for the first three sessions on the easiest level of the task (longest ISI) to see if the deterioration is still evident in the fourth test session. Study 2A examines whether naps can produce learning by lessening the number of sessions and increasing the nap duration to 1.5 hours to include both slow wave sleep and rapid eye movement sleep. Study 2B localizes learning demonstrated in the 1.5 hour naps to early visual cortex by testing for retinotopic specificity.
Exclusion criteria, Questionnaires, Sleep logs
All studies used the same subject criteria and requirements. Naïve subjects, obtained from an undergraduate study pool, were between the ages of 18 and 30, had never been diagnosed with any major mental illness, or sleep disorder, or suffered from head injuries and/or concussion, and they all had good or corrected vision. Subjects on psychoactive medication were excluded. Throughout testing subjects were not allowed any caffiene, and they were not allowed alcohol from the night before the first test until completion of the study. Subjects were asked to keep a sleep log of the hours they slept seven nights prior to testing and to sleep at least 7 hours on the eve of both test days. During the testing, subjects were asked to complete subjective rating forms. At the beginning of each test session, subjects rated their sleepiness levels on the Stanford Sleepiness Score. In study 1, subjects were tested on the texture discrimination task four times in one day, at 9AM, 12PM, 4PM, and 7PM. In study 2, subject are tested on the TDT twice on day one at 9am and 7pm, and in study 2A they are tested again on day two at 9am. Subjects are always tested on the same computer in a dimly lit, quiet room.
Subjects were randomly assigned to one of the nap conditions or to the no-nap control condition. Naps began at 2pm and recorded polysomnographically, with standard EEG, EOG, and EMG channels. Subjects were allowed to sleep until they had completed either a full half-hour, a full hour or a full 1.5 hours of sleep, depending on the study, of polysomnographically identified sleep and then were woken by the experimenter. Sleep stages were subsequently rescored offline. In study 1A, sleep was recorded during naps on two separate days, the TDT test day and a control day either one week before or one week after the test day, with the order balanced across subjects. The control and experimental naps in study A allowed for examination of changes in sleep quality due to testing condition
Study 1: Nature Neuroscience Publication, July 2002
Study 1A: The restorative effect of naps on perceptual deterioration.
In the first group of experiments, we examined a finding made earlier in a senior thesis by Harvard undergraduate Dan Luskin. He found that perceptual thresholds on the texture discrimination task do not remain stable when subjects are tested repeatedly within the same day, instead thresholds actually increase. This surprising deterioration in performance was the perfect setting for examining how daytime naps might influence perceptual performance. My first study tested participants on the task four times in one day (9AM, 12PM, 4PM, and 7PM). Control subjects (N=10) demonstrated a 52% slowing in perceptual processing across the four test sessions (Figure 3, filled circles) (p = .0003, repeated measures ANOVA and post-hoc tests). Thus, with each successive session, subjects needed increasingly longer exposures to the stimuli to reliably identify targets. This deterioration in performance was seen despite the fact that all testing was done within 12 hrs of morning awakenings and without prior sleep deprivation, and thus when one would not normally expect to see cognitive impairment. Subjects reported an average of 6.92 ± 0.77 (std. dev.) hrs of sleep on the night prior to testing. The deterioration demonstrated by subjects was unexpected. We have not found any precedent in the literature on visual perceptual learning to explain such a phenomenon.
Since nocturnal sleep is known both to enhance alertness and to consolidate TDT learning (Gais et al, 2000; Stickgold, 2000a; Karni et al, 1991; Karni, 1994)3, we asked whether a mid-day nap might stop or even reverse the process of deterioration seen with repeated within-day testing.
Twenty subjects assigned to a long nap (60 min) or short nap (30 min) condition, performed the task four times across the day, with the addition of a nap at 2 PM, midway between the second and third test sessions. As predicted, napping significantly affected subsequent performance (p = .001, group by session interaction, mixed-model ANOVA), with 30 min naps preventing the normal deterioration seen during sessions 3 and 4 (Fig. 3, open circles), and 60 min naps reversing the deterioration evident in the second session (Fig. 3, gray triangles). Thus, while controls showed a 14.1 ms increase in threshold between the second and third sessions, the short nap group showed no change ( .27).
Although these results demonstrate a clear advantage of nappers over non-nappers, there are many possible reasons for this performance reversal in the nap group. Perhaps it was not sleep, per se, that was essential for improving performance. Maybe just taking a rest without actually sleeping would be enough to restore performance. Further, subjects in the nap condition may have been psychologically more motivated simply because they were allowed to nap and therefore performed better. Alternatively, the control subjects, who were tested in an hour-long task, four times in one day, without a nap, became generally less motivated and less willing to perform well resulting in deteriorated performance. The next two experiments address these two possibilities and support the hypothesis that sleep and not quiet rest with eyes closed is essential for reversing the deterioration, and that increasing the control subject motivation by offering monetary incentive does not eliminate the deterioration. Both studies were conducted with Neha Pathak, a Harvard University undergraduate student doing her honor’s thesis with me.
Study 1B: The effect of quiet rest on perceptual deterioration.
In the quiet rest experiment, Neha Pathak and I asked whether the benefit noted was specifically due to sleep or whether simply having no visual input for an hour in a relaxed position was enough to benefit performance. To test this, subjects (N=9) repeated the long nap protocol, but the naps were replaced by quiet rest with blindfolding. Subjects listened to a book on tape. Wake-sleep state was continuously monitored physiologically to ensure maintained wakefulness. A new control group was tested with the same protocal but without the period of quiet rest. Despite the hour of rest without visual input, subjects showed continued performance decrements at 4 and 7 PM. Thresholds increased by an average of 29.0 ms between the second and fourth session (Fig.5: p < 0.05), a performance decrement similar (p = .31) to that of controls (Fig. 3). Thus, quiet rest failed to produce the improved performance seen with even 30 min naps.
Study 1C: The effect of increasing subject motivation on perceptual deterioration.
To test whether the performance decrements resulted from a decrease in motivation, subjects (N=10) were informed after their second session that their performance had worsened, and were told they would receive a cash bonus if they subsequently returned to their baseline performance. Despite this motivation, none of the subjects regained baseline performance during the third or fourth sessions, and mean thresholds were 32.2 ms slower on the fourth session compared to the first (Fig, 6: p = 0.001), a decrement nearly identical to the 40 ms seen in controls.
Study 1D: Retinotopic-specificity of perceptual deterioration.
Several mechanisms might underlie this deterioration. One would be a generalized fatigue effect, mediated by a decrease in alertness or attentional resources. Alternatively, specific neural networks in visual cortical areas may gradually become saturated with information through repeated testing, preventing further perceptual processing. This would be seen behaviorally as a training-specific deterioration in perceptual processing. While the findings reported above are consistent with both models for performance deterioration, the generalized fatigue hypothesis would further predict that decrements in performance should be widespread and largely task-independent. In contrast, our hypothesis of a training-specific deterioration predicts that the performance decrements should be restricted to behaviors mediated by the specific neural networks previously involved in processing the target stimuli. Since learning of the TDT does not transfer to untrained portions of the visual field (Karni and Sagi 1991), no training-specific deterioration should be seen if stimuli are presented to an untrained region of the visual cortex, and hence task performance should be normal.
To test this hypothesis, 24 subjects were trained and tested four times on one day, but with the target stimuli switched to the contralateral visual field for the final test. Performance of the switch group did not differ significantly from the control group across the first three sessions, but, unlike the control group, the switch group showed significant recovery in the fourth session (Fig. 7) (p = .002, ANOVA group x session interaction and post hoc test on the fourth session). Since performance during the switch condition was not significantly worse than during the first session, the behavioral deterioration observed in the trained visual quadrant did not transfer to the untrained contralateral quadrant. These results provide strong support for the training-specific deterioration hypothesis, and are contrary to the predictions of the generalized fatigue hypothesis.
Further evidence against the generalized fatigue hypothesis comes from the dissociation between improved performance and subjective levels of sleepiness (Fig. 8). If the steady decrease in performance across the day in the control group resulted from a general fatigue effect, then one should see a parallel increase in reported sleepiness. But no such increase was seen, and mean levels of subjective sleepiness on the first and last tests were identical (p = .45, repeated measures ANOVA). Similarly, the switch group showed no significant change in sleepiness across sessions (p = 0.49, repeated measures ANOVA). In contrast, sleepiness decreased from the first to the last session in the nap groups (p < .03, ANOVA and post-hoc tests). Thus, the switch group showed the same improvement in performance seen in the nap groups, but without a similar decrease in sleepiness, and the same maintenance in sleepiness levels as the controls but without the decrease in performance.
Data obtained from the switch group also eliminated another possible explanation, that of a strictly circadian effect. While the control data could be explained as a circadian rather than repetition effect, the fact that shifting the stimulus to the contralateral visual field for the last session reversed this decrease challenged this hypothesis. Thus, when subjects were tested at 7 PM with stimuli in an untrained region of visual space, they performed as well as they had at 9 AM that morning.
Study 1E: The effect of decreasing task difficulty on perceptual deterioration.
As this appears to be the first report of such a perceptual decline with repeated practice, we were interested in studying the limits of this phenomenon. This research was conducted with Alicia Levin as part of her honors thesis. To test whether the performance decrement resulted specifically from exposure to more difficult trials, with I.S.I.’s near or below threshold, subjects (N=10) were tested across the day in four sessions, with all blocks in the first three sessions run at the longest I.S.I. (i.e., 400 ms). When subjects were then tested in the fourth session with the standard 25 blocks of decreasing I.S.I’s, their performance did not significantly differ from that of the controls on their final session (p = 0.50). Thus, despite the decrease in task difficulty during the first three sessions, subjects appear to have shown no lessening of the performance decrement.
Mini-Summary of Study 1
In study 1, we showed that perceptual performance declined on the TDT with repeated, within-day training. In the context of this deterioration, we found that (i) a daytime nap, but not an equivalent period of rest without visual input, reversed the deterioration, (ii) the deterioration was retinotopically specific and (iii) neither an increase in subject motivation nor a decrease in task difficulty improved performance.
Discussion
Two learning components have been shown to occur with TDT testing, a fast, within-session component and a slow, sleep-dependent component (Karni, 1991). The present study identifies a third consequence of TDT training, that with repeated, same-day training, thresholds for texture discrimination do not remain stable, but instead increase. Such deterioration has not been reported with repeated same-day testing on other visual tasks. The nature of the task may contribute to this difference, such as whether the task measures vernier acuity (Bettina, Levi et al. 1995), resolution acuity (Bettina, Levi et al. 1995), or texture discrimination (Karni and Sagi 1991), whether stimuli are presented foveally (Fahle and Edelman 1993; Fahle 1994), parafoveally (Bettina, Levi et al. 1995), or at more peripheral eccentricities (Karni and Sagi 1991), and whether stimulus presentations are long (100 – 150 ms) (Fahle and Edelman 1993; Fahle 1994) or short (17 ms) (Karni and Sagi 1991). Furthermore, a number of perceptual learning protocols train subjects across days (Ball and Sekuler 1987; Fahle and Edelman 1993; Fahle 1994; Bettina, Levi et al. 1995; Watanabe, Nanez et al. 2001) rather than within-day, as in the current study, making it unclear whether fast or slow learning is occurring. With these caveats in mind, the present study shows that some forms of neural plasticity that require sleep for subsequent consolidation and improvement of perception may actually hinder further performance prior to sleep.
The findings that there is a normal decline in TDT performance across the day, which is dependent on repeated exposure to the task, is specific to previously trained regions of visual space, and can be reversed by napping, have implications. First, since circadian influences have been ruled out, the performance decline must result from specific neuronal changes induced by the initial testing period. Second, since brain regions involved in higher levels of visual processing lack retinotopic specificity, the critically affected neurons are most likely located in early visual processing areas. Finally, since the performance decrement is reversed by a nap, these initially affected neural networks must be further altered during napping in order to reverse the performance decrement.
One possible explanation for the performance decrement seen across the day is that the deterioration is actually a direct consequence of a mechanism that serves to preserve information previously processed, but not yet consolidated into memory by subsequent sleep. As this hypothesized limited capacity mechanism becomes saturated with task-specific information, the local neural network’s ability to process on-line information during task performance worsens, resulting in the performance decrement. Several findings support this relationship between information processing and the performance decrement. First, the retinotopic specificity of the performance decrement is in agreement with previous findings that suggest that learning of the TDT is similarly retinotopic (Karni and Sagi 1991) and dependent on early stages of visual processing (Karni and Sagi 1991; Karni, Weisberg et al. 1995). Second, post-training sleep is known to be critical for stabilization and consolidation of TDT learning (Gais, Plihal et al. 2000; Stickgold, James et al. 2000; Stickgold, Whidbee et al. 2000), and we now have shown that a 60 minute nap reverses the performance decrement.
While we cannot exclude a function for REM in this process, it seems most likely that SWS plays the dominant role in the reversal of deterioration. Though both SWS and REM have previously been implicated in the nocturnal, sleep-dependent consolidation and improvement on this task (Gais, Plihal et al. 2000; Stickgold, Whidbee et al. 2000), the REM-critical period is only observed 4 hrs after the SWS-dependent period has ended (Stickgold, Whidbee et al. 2000), well beyond the timeframe of these naps. Thus, we specifically hypothesize that, during SWS, mechanisms of cortical plasticity lead to secondary changes in the TDT-trained neural networks, producing the initial processing stage of the experience-dependent, long-term learning as well as a reversal of the performance decrement. Roles for SWS in memory consolidation have been proposed by others (Contreras, Destexhe et al. 1997; Plihal and Born 1997; Destexhe, Contreras et al. 1999; Gais, Plihal et al. 2000; Sejnowski and Destexhe 2000).
This example of a training-induced deterioration in performance has several additional implications. First, it indicates that the cognitive benefits of sleep can be studied over a very short time period and do not require sleep deprivation or overnight sessions of sleep. This provides a more favorable set of conditions to study the role of sleep in information processing and performance. Second, it suggests that the psychological sensation of "burnout," described anecdotally as increased irritation and frustration along with decreased effectiveness following prolonged cognitive effort, may not reflect a general mental fatigue, but rather the need of specific overused local neural networks for the restorative benefits of sleep.
Study 2: Sleep-dependent learning: A nap is as good as a night (submitted)
Study 2A: Nap-dependent learning depends on SWS and REM.
In the previous study, we showed that napping can produce long-term amelioration of the within-day, retinotopically specific, decline in perceptual performance. We did not, however, find improvement in texture discrimination beyond baseline; in other words it is not clear whether a nap can facilitate learning. In the next group of experiments, the question of whether nap-dependent learning is possible is addressed.
There are three possible reasons why we did not demonstrate nap-dependent learning in the previous experimental design. One, learning may have been overshadowed by the large magnitude of perceptual deterioration. Two, sleep-dependent learning may require a combination of both SWS and REM. Three, only a whole night of sleep can lead to improvement on the task. Both SWS and REM sleep have been implicated in a two-stage model of consolidation, with overnight improvement being highly correlated with the product of percents of early night SWS and late night REM sleep (Stickgold, 2000b). In my first study, 60 min naps reversed the daytime deterioration, but they did not lead to any net improvement in performance across the day. These beneficial naps had an increase in slow wave sleep (SWS) compared to baseline naps, as well as a large, although insignificant, increase in REM sleep. One possible explanation is a lack of sufficient REM sleep within the nap. Indeed, most subjects showed no REM sleep during 60-min naps.
The next experiment addressed whether longer naps may be as effective for learning as nocturnal sleep. We investigated both whether deterioration masked post-nap learning, as well as whether lengthening the naps to include more REM sleep would induce learning. Three groups (60 min, 90 min, no-nap controls) were tested twice on day one (9AM and 7PM), and once on day two (9AM). Two nap groups (60 min and 90 min nap groups) took polysomnographically recorded naps at 2pm. A fourth 24-hour control group (tested only twice - at 9AM on day one and 9AM on day two) was included in order to be able to compare performance of our experimental groups to that of a traditional nocturnal sleep-dependent learning group. The experimental design is illustrated in Figure 9. This study allowed for the following comparisons.
1. Nap-dependent vs. nocturnal sleep-dependent learning: we compared post-nap performance of the nappers to post-nocturnal sleep improvement of the no-nap controls and of the 24-hr controls.
2. The effect of a nap on over-night improvement: we compared levels of learning in nappers at test two and test three.
3. The effect of within-day deterioration on next day learning: we compared changes in performance of the no-nap controls at test two and test three.
We now report that sleep-dependent learning of a texture discrimination task can be accomplished by brief (60- to 90-minute) naps containing both slow wave sleep and rapid eye movement sleep. This nap-dependent learning appears indistinguishable from that previous reported for an 8-hr night of sleep in terms of magnitude, sleep stage dependency, and retinotopic specificity, and is additive to subsequent sleep-dependent improvement, such that performance over 24 hr shows as much learning as is normally seen after twice that length of time. Thus, from the perspective of behavioral improvement, a nap is as good as a night of sleep for learning on a perceptual task.
Subjects were trained on the texture discrimination task at 9AM and retested at 7PM that evening and then again at 9AM the next morning. Two experimental groups took naps at 2PM averaging 59 min (59.3 ± 6.4[s.d.]) and, 96 min (96.3 ± 6.3[s.d.]), while control subjects did not nap. Improvement was measured as a decrease in threshold from each subject’s 9AM baseline threshold. Table 3 shows the minutes in SWS and REM for the 60 and 90 min nap groups.
Control subjects showed the expected deterioration in performance at 7PM (-13.7 ms; p=0.06, Fig.10 left), and performed significantly worse than the nap groups (p=0.02). The deterioration in performance from training to first retest in the no-nap group, measured over an 9 hr interval, was similar to that seen over just a 2 hr interval in a prior study (-13.8 ms; p>0.9 (Mednick, 2002)). We consider this evidence that stimulus exposure rather than inter-test interval produces TDT deterioration, and that sleep, rather than time, is required to reverse this perceptual deterioration.
|Table 3: SWS and REM minutes of 60 and 90 minute nap |SWS (min) |REM (min) | (Threshold (ms) |
|groups | | | |
|60 min naps |20.2 ± 2.0 |4.2 ± 1.1 |4.4 ± 2.2 |
|90 min naps |47.2 ± 5.8 |25.6 ± 4.1 |8.4 ± 2.8 |
When tested at 7PM, the 90 min nappers showed significant improvement (8.4 ms, p=0.008), while the 60 min nappers showed marginal improvement (4.4 ms., p = 0.07). Mindful of our hypothesis that both SWS and REM may be necessary for learning in naps, we divided the 60-min nap group, all of whom had SWS, into subjects with and without REM and found that 60-min naps with both SWS and REM produced significant improvement (10.0 ms, p = 0.004, Fig 9 right, thick stripes). In contrast, 60-min naps with SWS but not REM showed no improvement (-1.1 ms, p = 0.72, Fig 10 center, thick stripes) and significantly less than seen in the SWS + REM group (p = 0.01). The 90-min nap group showed similar results (Fig. 10, thin stripes). When subjects with SWS + REM naps from the 60- and 90-min groups were combined, improvement correlated significantly with the product of the amount of SWS and REM sleep (r=0.37, p=0.01), in agreement with findings for nocturnal sleep (Stickgold, 2000b). In addition, the amount of improvement did not differ significantly from that previously reported (Stickgold, 2000a) for overnight improvement (9.7 vs. 11.9 ms; p=0.5). Thus, nap-dependent improvement showed the same magnitude and sleep-stage dependency as did overnight improvement. The finding that naps with SWS but not REM could reverse the deterioration but not produce actual improvement, while naps with SWS and REM were able to produce both, suggests that SWS may serve to stabilize and REM to improve performance.
Nap-dependent improvement was not at the expense of subsequent nocturnal improvement. On the contrary, when subjects were retested the next morning, the 90-min nap group showed an additional 9.7 ms of improvement (total=18.1 ms, p0.4).
The nap group actually showed 50% more improvement over a period of 24 hr than the 24-hr control group (18.1 ms vs. 11.8, p=.07). We compared 24-hr improvement in the nap group to previously published data of a 48-hr control group (retested after two nights of sleep, from Stickgold et al, 2000a). Indeed, the nap group showed equivalent levels of improvment as the 48-hr control group (Fig 11B; 18.1 ms vs. 17.5 ms; p>0.99). Taken together, these findings indicate that a 90 min nap can produce as much improvement as a night of sleep, and a nap followed by a night of sleep can provide as much benefit as two nights of sleep.
Study 2B: Retinotopic specificity of nap-dependent learning.
The post-nap improvement demonstrated in Study 2A is suspiciously similar to learning demonstrated after a night of sleep. We show that without both SWS and REM no improvement occurs, and learning is dependent on the product of the percentages of SWS and REM in the nap. But what is the specificity of the nap-dependent learning? At what level of cortical processing does the learning occur (i.e. sensory area or whole organism). Nocturnal sleep-dependent learning (on the TDT) represents plasticity in primary visual cortex (Karni et al, 1994). The same area of cortical processing underlies perceptual deterioration due to repeated within-day TDT testing (Study D: Mednick, 2002). Here we examine whether post-nap improvement also fits the specificity criterion for perceptual learning by testing whether the learning is specific to primary visual cortex.
As in Study 1D, we examined retinotopic specificity by having a Nap Switch group (N=6) test on the TDT in one quadrant at 9AM, nap for 1.5 hrs with PSG recording at 2PM, and then retest in the opposite quadrant at 7PM. If post-nap improvement is ubiquitous and seen at every level of processing then learning will generalize to other untrained areas of the visual field, but, if the improvement does not transfer to the opposite quadrant then we have shown that the post-nap improvement is retinotopically specific (i.e. occurs in primary visual cortex).
Indeed, eventhough the naps contained both SWS and REM, when the subjects were tested in an untrained quadrant at the second test, performance of the Nap Switch group did not improve significantly (p>.2). Figure 12 shows that subjects performance did not show performance benefits in the opposite qundrant compared to the 90min nap group. It is a mystery to us why subjects appear to show decreases in performance (4 out of the 5 subjects averaged -28.6ms (s.e. 4.48) decrease in performance on the 7PM test). Further research is necessary to investigate this question. But we have clearly shown that, as with overnight sleep-dependent improvement, nap-dependent improvement is retinotopically specific.
These findings demonstrate that naps can lead to improved performance on a texture discrimination task similar to previously reported overnight learning in terms of magnitude, retinotopic specificity and dependence on both SWS and REM. In addition, nap-dependent improvement significantly increases the amount of improvement which can develop over 24 hrs. Thus, a nap can not only ameliorate experience-dependent, perceptual deterioration, but also facilitate the learning process that results from an hour spent training on a visual texture discrimination task.
GENERAL CONCLUSIONS
This final discussion summarizes the findings from my dissertation research, discusses the relevance of the texture discrimination task for learning in general, and describes evidence of similar findings in other research. I will also consider how these findings of deterioration, restoration and learning fit into current models in perceptual learning.
Summary
More and more evidence is accumulating for the essential role of sleep for slowly developing perceptual learning. Previous studies have established that improvement on this texture discrimination task requires an extensive period of post-training nocturnal sleep, and the improvement is dependent on both early SWS and late REM sleep (Stickgold, 2000a; Karni et al, 1994; Gais et al, 2000). In this dissertation, we demonstrate another route to sleep dependent perceptual learning via the daytime nap.
We also report on a phenomenon resulting from repeated exposure to a visual texture discrimination task, that repeated within-day testing without an inter-test nap causes performance to worsen significantly in a retinotopic fashion with each successive test session. The deterioration does not diminish with time (up to 9 hrs), but only with sleep. The nap-dependent improvement varies depending on whether subjects sleep for a half hour, an hour, or an hour and a half. We found that after a half hour, performance that had deteriorated with repeated testing stabilized for two subsequent testing sessions. After an hour of sleep, deteriorated performance returned to baseline, but subjects did not improve on the TDT beyond their initial baseline levels. With an hour and a half nap, learning was demonstrated in significantly increased scores compared to baseline.
When we closely analyzed the improvement in the 60 min naps, we found that the lack of learning may have been due to the lack of REM sleep in the naps as well as the large amount of deterioration caused by multiple testing. Indeed, when the number of testing sessions was reduced from four to two, 60-min nappers showed marginally significant learning. Perhaps more important is that when the 60 min nap group was divided into subjects whose naps contained REM and those whose did not contain REM, we found that naps with REM led to significant learning. Performance following naps without REM, however, did not showed learning. In fact, the amount of learning demonstrated by the 60-min REM nap group was identical to learning shown in the 90-min nap group. A previous study reported that increased SWS and REM was correlated with increased learning (Stickgold et al, 2000b), it is therefore curious why naps of 90-min did not facilitate more learning than naps of 60-min. One possible explanation is that nocturnal sleep-dependent learning requires the extended arch of the night to facilitate learning, while only one sleep cycle is required of a nap.
We also found that the post-nap improvement was significantly correlated with the quality of sleep (the amount of SWS and REM) rather than the actually minutes slept. Twenty-four hours after the initial test time, 90 min nappers performed better than subjects who didn’t nap, in fact, they performed as well as subjects who showed enhanced improvement at a test 48 hrs after their initial training. These results suggest that including a nap in a normal sleep cycle may double the benefit of sleep for learning.
Relevance of texture discrimination for learning
It is our hope that we will be able to generalize our results across many tasks within the same category. One difficulty, however, in studying perceptual learning is that there are subtle differences in learning depending on the type of task utilized. This variability makes it difficult to generalize particular findings across a wide range of psychophysical tasks. Some learning requires sleep for consolidation (Karni and Sagi 1991), whereas other learning occurs over minutes (Fahle, Edelman et al. 1995). The magnitude and specificity of learning varies greatly between subjects (Beard et al, 1995), as well as between tasks. For example, variation in learning can depend on whether learning is tested in the fovea or parafovea (Beard et al, 1995), which type of stimulus is used (Beard et al, 1995), how wide the range of stimuli trained (Liu and Vaina 1998), and whether subjects are initially trained on easy or difficult test trials (Ahissar and Hochstein 2000). Such high variability presents concerns for interpreting our results in a larger context.
Our subjects also showed variability on the deterioration effect in the texture discrimination task, such as in performance decreases at the fourth test session in studies A & D (-45ms vs. -22ms, respectively). Such high variability could be due to many factors such as time of year of testing (fluctuations in the amount of work and stress across the semester may contribute to differences in level of deterioration), visual interference (amount of visual experience outside of the lab during the day of testing), nutrition of subjects (amount of sugar and protein in diet and caffeine dependence), individual differences in information capacity, and many other possible factors. Studying these phenomena in the environment of an in-patient sleep laboratory would control for the above factors, which may contribute to variability, as well as allowing for close examination of the role of circadian factors in learning and deteioration.
Alternative explanations for the cause of perceptual deterioration have been proposed. Prior work on reactive inhibition (Hull, 1943) report deterioration after repeated practice on simple repetitive tasks, such as letter cancellation, detecting differences in simple shapes, or adding three digits. These studies show that subjects show fluctuation in reaction time that are not related to the task itself, but rather to the subject performing the task (Smit et al, 1995). Reactive inhibition hypothesizes that these fluctuation are driven by a natural and constantly increasing inclination to switch from the task at hand to something else. This reactive inhibition closely parallels the concept of "fatigue" or "satiation" and explains a possible underlying mechanism in figural reversals, in which a figure, such as the Necker cube, switches between two perceptual forms with constant viewing. Researchers studying reactive inhibition explain this tendency with probabilistic models in which the brain is constantly poised to make the decision to switch or stay on a task.
In contrast to the reactive inhibition model that appears to propose a “boredom” mechanism in the brain, we show deterioration that builds only in a local neural network. This deterioration is not rooted in psychological explanations of boredom or fatigue as we have shown that increases in motivation do not ameliorate deterioration, and subjective sleepiness can be inversely correlated with performance. This network, however, can be relieved by switching to a different neural group or with sleep.
In a different example of deterioration, musicians and dancers anecdotally claim that after continuous practice on a difficult piece, rather than improve throughout the day, their performance worsens. In fact, some extremely dedicated musicians can actually suffer permanent damage to their sensorimotor abilities after extensive practice. In less severe cases, these musicians note great improvement on the same piece the following day. It is possible that this anecdotal deterioration is similar to the build-up of deterioration seen on the texture discrimination task. Further research into these questions would be helpful.
In this dissertation, we have not attempted to generalize the findings of deterioration or nap-dependent learning to other perceptual tasks. This makes it difficult to consider the extent to which these phenomena occur throughout visual and other sensory processing. There are, however, some indications in the perceptual learning literature that these phenomena may be relevant across a wider range of psychophysical tasks.
Hints of deterioration and sleep-dependence in perceptual learning studies
In most learning situations, it would be important to know whether the learning is sleep-dependent and whether repeated within-day testing might lead to deteriorated performance. This knowledge would not only help researchers in memory and learning but it would be helpful for students learning a new language, optimizing training schedules in the work place and for musicians and dancers, and many other applications. Although most perceptual learning studies do not traditionally report on the effects of over-practice or on the benefit of sleep for learning, indirect evidence for the benefit of sleep in perceptual learning can be found across a variety of studies.
Dosher and Lu studied retinotopic specific improvement on a task similar to the texture discrimination task by testing subjects 16 times across 8 days (Dosher and Lu, 1999). In their results, they mention that improvement was only found when subject were tested on successive days and that no improvement occurred when subjects were tested twice in one day. There is no mention of possible deleterious effects of repeated within-day testing.
Although methodological decisions regarding testing schedules are often not discussed in the literature, it appears that researchers are at least implicitly aware of the role of slow processes in learning. For example, Ball & Sekuler report on direction-specific motion learning which develops gradually across a ten day testing schedule (Ball and Sekuler 1987). The researchers do not, however, explain why they test subjects across days instead of using a simpler, within-day retest schedule. As another example, researchers at the University of Southern California studying visual motion learning have a policy only to test subjects on different days because they found that performance deteriorated with repeated, within-day testing on their motion discrimination task (Wilson Chu, personal communication, 2002).
In a meta-analysis of the rate and magnitude of learning across many different psychophysical tasks, ranging from spatial frequency and orientation discrimination to high-level object and facial recognition tasks, Fine and Jacobs limited their analysis to slow learning that occurs across a number of sessions noting “the role of sleep in consolidating learning (e.g. Karni, Tanne et al, 1994)”, and to avoid “possible fatigue effects” they “excluded studies where significantly more than an hour of training was carried out per day.” Although the above examples may not be considered direct evidence of deterioration and sleep-dependent learning, these studies indicate an awareness in the literature of an inherent limit to information processing and the benefit of 24 hr inter-session testing schedules.
Current models of perceptual learning
To understand sleep-dependent perceptual learning we can begin by looking at current models of perceptual learning. An important debate in the current perceptual learning literature is whether learning represents fine tuning of orientation-specific channels (V1 neurons) or plasticity in the weighting of inputs from basic visual mechanisms to decision. As proponents of the latter argument, Dosher and Lu propose that learning “primarily serves to select or strengthen the appropriate channel and prune or reduce inputs from irrelevant channels (Dosher et al. 1999).” They conclude that learning represents greater weighting of connections between visual channels most closely tuned to the target stimulus. This group of connections forms a learned category that is maintained and strengthened, while input from other channels is reduced or eliminated (Jenkins, Merzenich et al. 1990; Recanzone, Merzenich et al. 1992; Recanzone, Schreiner et al. 1993; Kilgard et al, 1998).
In contrast, research consistent with an early tuning mechanism can be found in work examining the receptive field properties of neurons in V1 and V2 representing trained or untrained locations (Ghose et al, 2002). After training, slightly fewer neurons were found whose optimal orientation was near the trained orientation. This resulted in a small but significant decrease in the V1 population response to the trained orientation at the trained location. The authors concluded that these results represented the fine tuning of orientation specific neurons.
Dosher and Lu, however, argue that a change in the earliest visual mechanisms due to learning on one task would necessarily affect learning on another or a previously learned task. In addition, generalization of learning is dependent on task requirements, such as level of difficulty (Ahissar and Hochstein 2000) and the range of stimuli trained (Liu and Vaina 1998). This suggests that some middle level process may be regulating this flexibility in behavior of individual and groups of neurons. Thus, plasticity may not occur at the most basic visual mechanisms themselves nor at the level of general cognitive strategy or decision mechanism, but at an intermediary level connecting basic and high level structures. Long range horizontal connections in the visual cortex and feedback connections from higher order cortical areas have been implicated.
Ahissar and Hochstein (2000) propose a model that aims to define rules of generalization and specificity of perceptual learning by introducing control pathways between low-level visual channels and high level cognitive processes. This model attempts to explain variability in specificity of learning by demonstrating an inverse relationship between task difficulty and generalizability. The more difficult the task the less generalized the learning. Their reverse hierarchy model proposes that learning on an attentionally “easy” trial (e.g., pop-out trial) is controlled by higher cognitive processes. As the task increases in difficulty, attention is focused on lower levels of processing where neurons have narrowly tuned receptive fields and thus learn highly specific information. Liu (2000) describes a similar predictive learning model and adds that with an increase in range of stimuli at training, there is an increase in generalizability of learning. An example of this rule may be seen in motion learning. Ball and Sekuler tested subjects on one motion direction and found direction-specific learning (1987), whereas Liu (2000) tested subjects on multiple motion directions and found that learning transferred to untested directions of motion.
For the texture discrimination task, the characteristics of learning appear to correspond to both the reverse hierarchy model and the rule of generalization. Subjects require a training block of 50 trials set at the easiest ISI (400ms) before they are able to see and correctly identify the peripheral target. Also, subjects trained on a limited range of targets (two orientations in one visual field), do not show transfer of either learning or deterioration. However, an important aspect of learning that these models do not address is the difference between sleep-dependent and sleep-independent learning. The role of sleep illuminates an important part of the mechanism of plasticity by contributing information on the effect of specific sleep stages on neuronal behavior.
In this dissertation, we have traced the functioning of two phenomena, perceptual learning and deterioration, to the same area of primary visual cortex. We have found that these phenomena are both affected by sleep. We have shown that with repeated within-day testing on the TDT, the neural network processing the textured target performs increasingly poorly, and is only ameliorated with sleep. The same level of deterioration was found whether testing sessions were separated by two hours or nine hours, indicating that spontaneous restoration of the neural network did not occur across a nine hour period of waking. Restoration failed to occur with quiet rest, increased motivation, or decreased task difficulty. We found that naps containing SWS contribute to the restoration of the neural network, and that naps rich in SWS and REM facilitate plasticity of the visual cortex underlying learning.
With the marriage of sleep and learning, interesting new questions arise. For example, we can now ask about the independence of the two phenomena of learning and deterioration. Are learning and deterioration separable phenomena that develop independently of each other with different underlying mechanisms? Alternatively, do learning and deterioration lie on opposite ends of the same continuum, such as an experience-dependent continuum that describes neuronal behavior as it is modualted by sleep. Another interesting issue is the role of attention in learning and deterioration. Recent research in attention using functional magnetic resonance imaging shows that cortical areas as early as V1 are modulated by shifts in the allocation of attention (Gandhi, Heeger & Boynton, 1999). An exciting addition to this dissertation research would be to assess the influence of attention on these changes in local neural network caused by experience and sleep. My post-doctoral research will focus on this issue.
In conclusion, we have demonstrated a new route to sleep-dependent learning via a daytime nap. We hope that these data will inform research in learning and memory of a mechanism for slow-developing consolidation, and inform researchers in visual perception of ways to optimize data collection, as well as legitimize universal napping across the globe.
Ahissar, M . and S. Hochstein (1993). “Attentional control of early perceptual learning.” Proc Natl Acad Sci U S A 90(12): 5718-22.
Ahissar, M. and S. Hochstein (1997). “Task difficulty and the specificity of perceptual learning.” Nature 387(6631): 401-6.
Ahissar, M. and S. Hochstein (2000). “The spread of attention and learning in feature search: effects of target distribution and task difficulty.” Vision Res 40(10-12): 1349-64.
Ball, K. and R. Sekuler (1987). “Direction-specific improvement in motion discrimination.” Vision Res 27(6): 953-65.
Beard BL, L. D., Reich LN. (1995). “Perceptual learning in parafoveal vision.” Vision Res. Jun;35(12):1679-90.
Bettina, B. L., D. M. Levi, et al. (1995). Vision Research 34: 1679-1690.
Blakemore C, C. G. (1970 Oct 31;). “Development of the brain depends on the visual environment.” Nature. 228(270):477-8.
Blakemore C, V. S. R. (1975 Jul;). “Innate and environmental factors in the development of the kitten's visual cortex.” J Physiol. 248(3):663-716.
Blasdel (1992). “Differential imaging of ocular dominance and orientation selectivity in monkey striate cortex.” J Neurosci. Aug;12(8):3115-38.
Bonnet M. H., D. L. Arand, Physiology and Behavior 56, 103-109 (1994).
Bonnet M. H. , Sleep 14, 307-315 (1991).
Broadbent, D. E. (1958). Perception and communication. New York:, Pergamon,.
Burkhalter (1993). “Development of forward and feedback connections between areas V1 and V2 of human visual cortex.” Cereb Cortex. Sep-Oct;3(5):476-87.
Contreras, D., A. Destexhe, et al. (1997). “Intracellular and computational characterization of the intracortical inhibitory control of synchronized thalamic inputs in vivo.” Journal of Neurophysiology 78: 335-350.
Crist, R. E., M. K. Kapadia, et al. (1997). “Perceptual learning of spatial localization: specificity for orientation, position, and context.” J Neurophysiol 78(6): 2889-94.
Crist, R. E., W. Li, et al. (2001). “Learning to see: experience and attention in primary visual cortex.” Nat Neurosci 4(5): 519-25.
Dalva MB, K. L. (1994). “Rearrangements of synaptic connections in visual cortex revealed by laser photostimulation.” Science. Jul 8 265(5169):255-8.
Dinges D. F. , R. J. Broughton, Eds., Sleep and Alertness: Chronobiological, Behavioral, and Medical Aspects of Napping (Raven Press, New York, 1989).
Dinges, D. F. (1992). Adult napping and its effects on ability to function. Why we nap: Evolution, chronobiology, and functions of polyphasic and ultrashort sleep. C. Stampi. Boston, MA, USA, Birkhaeuser. XXII: 279 pp.
Dosher BA, L. Z. (1999). “Mechanisms of perceptual learning.” Vision Research 39 3197-3221.
Fahle, M. (1994). “Human pattern recognition: parallel processing and perceptual learning.” Perception 23: 411-427.
Fahle, M. and S. Edelman (1993). “Long-term learning in vernier acuity: Effects of stimulus orientation, range and of feedback.” Vision Research 33: 397-412.
Fahle, M., S. Edelman, et al. (1995). “Fast perceptual learning in hyperacuity.” Vision Res 35(21): 3003-13.
Fiorentini, A. and N. Berardi (1980). “Perceptual learning specific for orientation and spatial frequency.” Nature 287(5777): 43-4.
Frank, M. G., N. P. Issa, et al. (2001). “Sleep enhances plasticity in the developing visual cortex.” Neuron 30: 275-287.
Gais, S., W. Plihal, et al. (2000). “Early sleep triggers memory for early visual discrimination skills.” Nature Neuroscience 3: 1335-1339.
Gandhi, S. P., D. J. Heeger, et al. (1999). “Spatial attention affects brain activity in human primary visual cortex.” Proc. Natl. Acad. Sci. USA 96: 3314-3319.
Gilbert C.D., Hirsh J.A., Wiesel TN. (1990). “Lateral interactions in visual cortex.” Cold Spring Harb Symp Quant Biol. ;55:663-77. Review.
Gilbert C.D., Sigman M., Crist R.E. (2001). “The neural basis of perceptual learning.” NEURON SEP 13 31(5):681-697.
Gilbert C.D., Wiesel. T.N. (1983). “Clustered intrinsic connections in cat visual cortex.” J Neurosci. May;3(5):1116-33.
Ghose, G. M., T. Yang, et al. (2002). “Physiological correlates of perceptual learning in monkey V1 and V2.” J Neurophysiol Apr;87(4):1867-88.
Graves, L., A. Pack, et al. (2001). “Sleep and memory: a molecular perspective.” Trends in Neurosciences 24: 237-243.
Harma, M., P. Knauth, et al. (1989). “Daytime napping and its effects on alertness and short-term memory performance in shiftworkers.” International Archives of Occupational and Environmental Health 61: 341-345.
Hasselmo, M. E. (1999). “Neuromodulation: acetylcholine and memory consolidation.” Trends in Cognitive Sciences 3: 351-359.
Herrera, D. G. and H. A. Robertson (1996). “Activation of c-fos in the brain.” Prog Neurobiol 50(2-3): 83-107.
Hobson, J. A. (1975). The sleep-dream cycle: a neurobiological rhythm. Pathobiology Annual 1975. H. L. Ioachim. New York, Appleton-Century-Crofts: 369-403.
Horne J. A. , L. A. Reyner, Psychophysiology 33, 306-309 (1996).
Hubel, D. H. and T. N. Wiesel (1968). “Receptive fields and functional architecture of monkey striate cortex.” J Physiol 195(1): 215-43.
Hubel, D. H. and T. N. Wiesel (1974). “Uniformity of monkey striate cortex: a parallel relationship between field size, scatter, and magnification factor.” J Comp Neurol 158(3): 295-305.
Hubel, D. H., T. N. Wiesel, et al. (1978). “Anatomical demonstration of orientation columns in macaque monkey.” J Comp Neurol 177(3): 361-80.
Hull, C. L. (1952). A behavior system: An introduction to behavior theory concerning the individual organism. New Haven, Conn., Yale University Press.
Jenkins, W. M., M. M. Merzenich, et al. (1990). “Neocortical representational dynamics in adult primates: implications for neuropsychology.” Neuropsychologia 28(6): 573-84.
Karni, A. and D. Sagi (1991). “Where practice makes perfect in texture discrimination: evidence for primary visual cortex plasticity.” Proceedings of the National Academy of Science of the United States of America 88: 4966-4970.
Karni, A. and D. Sagi (1993). “The time course of learning a visual skill.” Nature 365: 250-252.
Karni, A., D. Tanne, et al. (1994). “Dependence on REM sleep of overnight improvement of a perceptual skill [see comments].” Science 265(5172): 679-82.
Karni, A., J. Weisberg, et al. (1995). “Slow changes in primary and secondary visual cortex associated with perceptual skill learning: An fMRI study.” NeuroImage 3: S543.
Katz LC, C. E. (1992). “Development of local circuits in mammalian visual cortex.” Annu Rev Neurosci. 15:31-56. Review.
Kilgard MP, M. M. (1998). “Plasticity of temporal information processing in the primary auditory cortex.” Nat Neurosci. Dec;1(8):727-31.
Liu, Z. and L. M. Vaina (1998). “Simultaneous learning of motion discrimination in two directions.” Brain Res Cogn Brain Res 6(4): 347-9.
Liu Z, W. D. (2000). “Mechanisms of generalization in perceptual learning.” Vision Res. 40(1):97-109.
Maquet, P. (2001). “The role of sleep in learning and memory.” Science 294: 1048-1052.
Mednick, S. C., K. Nakayama, et al. (2002). “The restorative effect of naps on perceptual deterioration.” Nature Neuroscience 5(7): in press.
McGaugh, J. L., P. E. Gold, et al. (1979). Altering memory by electrical and chemical stimulation of the brain. Brain Mechanisms in Memory and Learning: From the Single Neuron.
Milner, B. (1962). Les troubles de la memoire accompagnant des lesions hippocampiques bilaterales. Physiologie de l'hippocampe. Paris, Centre National de la Recherche Scientifique: 257-272.
Mishkin, M., F. Vargha-Khadem, et al. (1998). “Amnesia and the organization of the hippocampal system.” Hippocampus 8(3): 212-6.
Plihal, W. and J. Born (1997). “Effects of early and late nocturnal sleep on declarative and procedural memory.” Journal of Cognitive Neuroscience 9(4): 534-547.
Recanzone, G. H., M. M. Merzenich, et al. (1992). “Topographic reorganization of the hand representation in cortical area 3b of owl monkeys trained in a frequency-discrimination task.” Journal of Neurophysiology 67: 1031-1056.
Recanzone, G. H., C. E. Schreiner, et al. (1993). “Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys.” Journal of Neuroscience 13: 87-103.
Rechtschaffen, A. and A. Kales (1968). A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects., Brain Information Service, University of California, Los Angeles.
Roffwarg, H. P., J. N. Muzio, et al. (1966). “Ontogenetic development of the human sleep-dream cycle.” Science 152: 604-619.
Rosa R. R. , Sleep 16, 727-735 (1993).
Rubin, N., K. Nakayama, et al. (1997). “Abrupt learning and retinal size specificity in illusory-contour perception.” Curr Biol 7(7): 461-7.
Schacter, D. L. (1985). “Priming of old and new knowledge in amnesic patients and normal subjects.” Ann N Y Acad Sci 444: 41-53.
Schacter, D. L. and E. Tulving (1994). Memory systems 1994. Cambridge, MA, MIT Press.
Schwartz, S., Maquet, P.,Frith, C., (2002). “Neural correlates of perceptual learning: a functional MRI study of visual texture discrimination.” Proc Natl Acad Sci U S A Dec 24;99(26):17137-42
Scoville, W. B. and B. Milner (1957). “Loss of recent memory after bilateral hippocampal lesions.” Journal of Neurology, Neurosurgery and Psychiatry 20: 11-21.
Sejnowski, T. J. and A. Destexhe (2000). “Why do we sleep?” Brain Research 886((Special Issue SI)): (S1) 208-223. Plasticity in Human Early Visual Areas Revealed by fMRI.” .
Smit, J. C. a. v. d. V., A.H.G.S. (1995). “Inhibition in Speed and Concentration Tests: The Poisson Inhibition Model.” Journal of Mathematical Psychology 39, 265-274.
Smith, C. and G. M. Rose (1996). “Evidence for a paradoxical sleep window for place learning in the Morris water maze.” Physiology and Behavior 59: 93-97.
Sperling, G. (1960). The information available in brief visual presentations., Psychological Monographs,.
Squire, L. R. (1994). Declarative and nondeclarative memory: multiple brain systems supporting learning and memory. Memory Systems 1994. D. L. Schacter and E. Tulving. Cambridge MA, MIT Press: 203-232.
Stein, M. (2001). Less fun, less sleep, more work: An American portrait, National Sleep Foundation.
Stickgold, R., J. A. Hobson, et al. (2001). “Sleep, learning and dreams: Off-line memory reprocessing.” Science 294: 1052-1057.
Stickgold, R., L. James, et al. (2000a). “Visual discrimination learning requires sleep after training.” Nat Neurosci 3(12): 1237-8. .
Stickgold, R., B. Schirmer, et al. (1998). “Visual discrimination learning: Both NREM and REM are required.” Sleep 21 (Supplement): 256.
Stickgold, R., D. Whidbee, et al. (2000b). “Visual discrimination task improvement: A multi-step process occurring during sleep.” J Cogn Neurosci 12(2): 246-54.
Takahashi M. , H. Arito, Sleep 23, 813-819 (2000).
Taub J. M. , International Journal of Neuroscience 9, 97-112 (1979).
Walker, M., T. Brakefield, et al. (2002). “Practice with sleep makes perfect: Sleep dependent motor skill learning.” Neuron 35(2): (in press)
Watanabe, T., J. E. Nanez, et al. (2001). “Perceptual learning without perception.” Nature 413: 844-848.
Weiskrantz, L. and E. K. Warrington (1979). “Conditioning in amnesic patients.” Neuropsychologia 17(2): 187-94.
Zohary, E., S. Celebrini, et al. (1994). “Neuronal plasticity that underlies improvement in perceptual performance.” Science 263(5151): 1289-92.
-----------------------
Figure 3: Recovery of performance with napping
Figure 6: Motivation group vs. Controls
Figure 5: Quiet Rest and Controls
*
*
Experimental
Baseline
Minutes
Figure 4: Sleep stages of baseline vs. experimental naps in long nappers
REM
SWS
S2
S1
35
30
25
20
15
10
5
0
[pic]
[pic]
[pic]
Figure 7: Retinotopic deterioration in texture discrimination. Both groups showed deterioration in session 3 (T3-T1), but in the fourth session (T4-T1), the controls deteriorated further while the switch group recovered to baseline levels of performance
15
10
5
0
-5
-10
-15
Figure 10: Same-day improvement scores in no-nap, 60' nap, and 90' nap groups, divided by presence or absence of REM. Left: No-Nap group shows deterioration at 7PM from baseline test at 9AM. Center: 60' nap with SWS and no REM shows no deterioration but no improvement. Right, thick stripes: 60' nap with SWS and REM shows significant improvement at 7PM. Right, thin stripes: 90' nap with SWS and REM shows improvement at 7PM. Only two subjects in the 90‘ nap group showed no REM.
[pic]
A. Day 2
Figure 1: EEG representation of sleep/wake stages and sleep architecture throughout the night (from Kandel, Schwatz, Jessel, 1998)
Table 2: Characteristics of short and long naps. The mean times spent in each sleep stage during short (30 min) and long (60 min) naps are presented as minutes ± s.e.m. S1 = Stage 1, S2 = Stage 2, SWS = slow wave sleep (Stages 3 and 4), REM = rapid eye movement sleep. (p): significance of unpaired t-test comparing times spent in stage in long and short naps; (Threshold: the difference between TDT thresholds on the second (pre-nap) and third (post-nap) tests, expressed as ms ± s.e.m.
Improvement (ms)
no nap
+SWS
+REM
(n=13)
(n=17)
+SWS-REM
(n=2)
(n=13)
-SWS
-REM
(n=28)
60’ nap
90’nap
No-nap Control
(24 hr)
Nap
A. Day 2
0
5
10
15
20
25
Improvement (ms)
Control
(48 hr)
B. Day 3
Figure 11. Performance changes for nappers and non-nappers. A: Performance changes 24 hr post-training, for the No-nap group, second retest at 9AM, the 24 hr group, first retest at 9AM, and the 90 min nap group, second retest at 9AM. Dotted line shows Nap improvement on day 1; B: Performance changes 48 hr post-training. 24 and 48 hr controls from (Stickgold, 2000a) .
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- and when necessary use words
- the necessary and proper clause
- uses of the necessary and proper clause
- technology and the impacts on society
- the necessary and proper clause quizlet
- where is the necessary and proper clause
- define the necessary and proper clause
- applications of the necessary and proper clause
- tips on getting to work on time
- the necessary and proper clause states that
- list the necessary life functions
- what does the necessary and proper clause