In terms of functional anatomy, where does learning occur when, for a basic visual discrimination task, performance improves with practice (perceptual learning)? We report remarkable long-term learning in a simple texture discrimination task where learning is specific for retinal input.This learning is (i) local (in a retinotopic sense), (i) orientation specific but asymmetric (it is specific for background but not for target-element orientation), and (ii) strongly monocular (there is little interocular transfer of learning). Our results suggest that learning involves experience-dependent changes at a level of the visual system where monocularity and the retinotopic organization of the visual input are still retained and where different orientations are processed separately.These results can be interpreted in terms of local plasticity induced by retinal input in early visual processing in human adults, presumably at the level of orientation-gradient sensitive cells in primary visual cortex.Early visual processing is believed to be "hard-wired" in adult mammals. Yet for some simple visual discrimination tasks, performance improves with practice-i.e., a genuine increase in sensitivity is induced by sensory experience (perceptual learning). Some instances of perceptual learning are specific for particular stimulus parameters and do not transfer when these are changed (1-4). This has interesting implications for understanding sensory processing because the dependence of learning on specific stimulus parameters could provide an effective probe to the functional architecture of the sensory system. We applied this logic to investigate the effects of training on simple texture discrimination and explored the possibility that strictly local, orientationsensitive mechanisms were selectively involved in the learning process.We were motivated by the observation that although recent theories of vision suggest that some texture segregations are accomplished very early in visual processing in an "automatic" (preattentive) way [i.e., without the need for higher level (attentive) recognition (5-8)], observers improve with practice even in simple, preattentive texture discrimination. Previous work referred to the "overlearning" of the limitations of "nonautomatic" attentive vision or difficult texture discriminations (6, 9). We were intrigued by the fact that observers seemed to improve even in simple automatic texture segregation. If learning is indeed local and orientation specific, it must presumably involve changes at an early, low-level processing stage where the retinotopic organization of visual input is retained and different orientations are handled separately (10, 11). A high degree of monocularity (i.e., learning that would not transfer from a trained to an untrained eye) would suggest that learning affected a level within the visual system where cells preferentially respond to input from one retina (monocular cells) (10, 11).

Behavioral and neurophysiological studies suggest that skill learning can be mediated by discrete, experience-driven changes within specific neural representations subserving the performance of the trained task. We have shown that a few minutes of daily practice on a sequential finger opposition task induced large, incremental performance gains over a few weeks of training. These gains did not generalize to the contralateral hand nor to a matched sequence of identical component movements, suggesting that a lateralized representation of the learned sequence of movements evolved through practice. This interpretation was supported by functional MRI data showing that a more extensive representation of the trained sequence emerged in primary motor cortex after 3 weeks of training. The imaging data, however, also indicated important changes occurring in primary motor cortex during the initial scanning sessions, which we proposed may ref lect the setting up of a task-specific motor processing routine. Here we provide behavioral and functional MRI data on experience-dependent changes induced by a limited amount of repetitions within the first imaging session. We show that this limited training experience can be sufficient to trigger performance gains that require time to become evident. We propose that skilled motor performance is acquired in several stages: "fast" learning, an initial, withinsession improvement phase, followed by a period of consolidation of several hours duration, and then "slow" learning, consisting of delayed, incremental gains in performance emerging after continued practice. This time course may ref lect basic mechanisms of neuronal plasticity in the adult brain that subserve the acquisition and retention of many different skills.The performance of many tasks improves, throughout life, with repetition and practice. Even in adulthood simple tasks such as reaching to a target or rapidly and accurately tapping a short sequence of finger movements, which appear, when mastered, to be effortlessly performed, often require extensive training before skilled performance develops. What changes occur in the adult brain when a new skill is acquired through practice? When, and after how much practice, do these changes occur? Functional reorganization of adult mammalian sensory and motor cortical representations has been found to occur in many different animal models of brain plasticity in the last two decades, advancing the idea that throughout life the functional properties of central nervous system neurons, as well as the neural circuitry within different brain areas, are malleable and retain a functionally significant degree of plasticity (e.g., refs. 1-4). These representational changes have been shown to be induced not only in response to lesions of peripheral or central sensory input or motor output pathways but also, in normal individuals, as a result of practice and experience. The advent of new brain imaging techniques, especially functional MRI (fMRI) (5), which allows repeated mapping of cor...

Several examples of experience-dependent perceptual improvement (perceptual learning) suggest that plasticity in specific neuronal loci could underlie the learning process. For a basic visual discrimination task (using an optimal stimulus for 'automatic' pre-attentive texture segregation), discrete retinal input-dependent changes within a very early stage in the stream of visual processing were indicated as the locus of a large and consistent learning effect. When do these changes occur? Here we report that except for a fast, rapidly saturating improvement early in the first practice session, performance was very stable within sessions. Indeed, observers showed little or no improvement until up to 8 hours after their last training session (latent phase). But large improvements occurred thereafter. Finally, there was almost no forgetting; what was gained was retained for at least 2-3 years. We conjecture that some types of perceptual experience trigger permanent neural changes in early processing stages of the adult visual system. These may take many hours to become functional.

Several paradigms of perceptual learning suggest that practice can trigger long-term, experience-dependent changes in the adult visual system of humans. As shown here, performance of a basic visual discrimination task improved after a normal night's sleep. Selective disruption of rapid eye movement (REM) sleep resulted in no performance gain during a comparable sleep interval, although non-REM slow-wave sleep disruption did not affect improvement. On the other hand, deprivation of REM sleep had no detrimental effects on the performance of a similar, but previously learned, task. These results indicate that a process of human memory consolidation, active during sleep, is strongly dependent on REM sleep.

Memory consolidation refers to the transformation over time of experience-dependent internal representations and their neurobiological underpinnings. The process is assumed to be embodied in synaptic and cellular modifications at brain circuits in which the memory is initially encoded and to proceed by recurrent reactivations, both during wakefulness and during sleep, culminating in the distribution of information to additional locales and integration of new information into existing knowledge. We present snapshots of our current knowledge and gaps in knowledge concerning the progress of consolidation over time and the cognitive architecture that supports it and shapes our long-term memories.

Several lines of evidence support the notion that the brain exhibits a significant degree of experience-dependent functional plasticity even in adulthood (1-10). This plasticity may underlie the acquisition and long-term retention of skills (procedural memory) (11-13). There is growing evidence, however, indicating that different brain areas are involved in the initial, compared with subsequent, phases of learning after practice in a given motor as well as nonmotor task (14-22). These practicerelated changes in the set of brain areas engaged by a repeating task were reported to occur mostly within a single session. However, recent behavioral data have shown that the acquisition of skilled performance occurs on a time scale of hours, days, and weeks (2, 5, 7, 11-13, 18, 21-24). A leading notion, the ''power law of practice,'' suggests that the evolution of skilled performance is determined solely by the number of task repetitions (1-5, 7, 11-13, 25, 26); there are, nevertheless, compelling indications that the passage of time is also an important factor in the acquisition of skills (2,24,(27)(28)(29).Based on behavioral and imaging studies, two distinct stages in skill acquisition were proposed: early, relating to withinsession improvement, and late, slow changes in performance that can be observed across several (daily) sessions of practice (1,2,5,7,10,12,13,28). To account for robust delayed gains in performance that emerged after a latent period of more than a few hours after a single-session training, the notion of an intermediate phase corresponding to the posttraining hours has been proposed (2, 7, 13, 27-31). The conjecture is that a process of memory consolidation requiring time to become effective (in terms of performance) can be triggered by the training experience under certain conditions and requires time to become effective (in terms of performance) in both perceptual and motor tasks [but also in the acquisition of cognitive skills (12)].Recent studies further suggest that sleep may contribute significantly to the development of the delayed gains in this type of learning (28,30,32).It is not clear, however, whether the effects of a single training session, with ample time given for the evolution of delayed gains, can be conceptualized as the unit of skill acquisition, i.e., that multisession training gain constitutes but the sum of incremental gains of a number of single sessions. An earlier study (2) showed that all gains in speed of performance after completing longterm training on a sequence of movements were highly restricted by the physical parameters of the training experience, with no transfer to the untrained hand or to different arrangements of the trained movement components comprising the sequence. Here we show that this remarkable specificity evolves in a stepwise manner both within and between sessions. Within a given session, large performance gains occurred only for newly introduced conditions irrespective of the absolute level of performance. Although after a single session qualitativ...

Two behavioral phenomena characterize human motor memory consolidation: diminishing susceptibility to interference by a subsequent experience and the emergence of delayed, offline gains in performance. A recent model proposes that the sleep-independent reduction in interference is followed by the sleep-dependent expression of offline gains. Here, using the finger-opposition sequence-learning task, we show that an interference experienced at 2 h, but not 8 h, following the initial training prevented the expression of delayed gains at 24 h post-training. However, a 90-min nap, immediately post-training, markedly reduced the susceptibility to interference, with robust delayed gains expressed overnight, despite interference at 2 h post-training. With no interference, a nap resulted in much earlier expression of delayed gains, within 8 h post-training. These results suggest that the evolution of robustness to interference and the evolution of delayed gains can coincide immediately post-training and that both effects reflect sleep-sensitive processes.

No abstract