Versatile perceptual learning of textures after variable exposures

Perceptual learning of 10-AFC texture identification is stimulus specific: after practice, identification accuracy drops substantially when textures are rotated 180°, reversed in contrast polarity, or when a novel set of textures is presented. Here we asked if perceptual learning occurs without any repetition of items during training, and whether exposure to greater stimulus variation during training influences transfer of learning. We trained three groups of subjects in a 10-AFC texture identification task on 2 days. The Standard group viewed a fixed set of 10 textures throughout training. The Variable group viewed 840 novel sets of textures. The Switch group viewed different fixed sets of 10 textures on Days 1 and 2. In all groups, transfer of learning was tested by using fixed sets of textures on Days 3 and 4 and having half of the subjects from each group switch to a novel set on Day 4. During training, the most learning was obtained by the Standard group, and gradual but significant learning was obtained by the other two groups. On Day 4, performance of the Standard group was adversely affected by a switch to novel textures, whereas performance of the Variable and Switch groups remained intact. Hence, slight but significant learning occurred without repetition of items during training, and stimulus specificity was influenced significantly by the type of training. Increasing stimulus variability by reducing the number of times stimuli are repeated during practice may cause subjects to adopt strategies that increase generalization of learning to new stimuli. Alternatively, presenting new stimuli on each trial may prevent subjects from adopting strategies that result in stimulus specific learning.     

S-cone discrimination for stimuli with spatial and temporal chromatic contrast

In the natural environment, color discriminations are made within a rich context of spatial and temporal variation. In classical laboratory methods for studying chromatic discrimination, there is typically a border between the test and adapting fields that introduces a spatial chromatic contrast signal. Typically, the roles of spatial and temporal contrast on chromatic discrimination are not assessed in the laboratory approach. In this study, S-cone discrimination was measured using stimulus paradigms that controlled the level of spatio-temporal S-cone contrast between the tests and adapting fields. The results indicate that S-cone discrimination of chromaticity differences between a pedestal and adapting surround is equivalent for stimuli containing spatial, temporal or spatial-and-temporal chromatic contrast between the test field and the surround. For a stimulus condition that did not contain spatial or temporal contrast, the visual system adapted to the pedestal instead of the surround. The data are interpreted in terms of a model consistent with primate koniocellular pathway physiology. The paradigms provide an approach for studying the effects of spatial and temporal contrast on discrimination in natural scenes.     

Training in contrast detection improves motion perception of sinewave gratings in amblyopia

PURPOSE. One critical concern about using perceptual learning to treat amblyopia is whether training with one particular stimulus and task generalizes to other stimuli and tasks. In the spatial domain, it has been found that the bandwidth of contrast sensitivity improvement is much broader in amblyopes than in normals. Because previous studies suggested the local motion deficits in amblyopia are explained by the spatial vision deficits, the hypothesis for this study was that training in the spatial domain could benefit motion perception of sinewave gratings. METHODS. Nine adult amblyopes (mean age, 22.1 ± 5.6 years) were trained in a contrast detection task in the amblyopic eye for 10 days. Visual acuity, spatial contrast sensitivity functions, and temporal modulation transfer functions (MTF) for sinewave motion detection and discrimination were measured for each eye before and after training. Eight adult amblyopes (mean age, 22.6 ± 6.7 years) served as control subjects. RESULTS. In the amblyopic eye, training improved (1) contrast sensitivity by 6.6 dB (or 113.8%) across spatial frequencies, with a bandwidth of 4.4 octaves; (2) sensitivity of motion detection and discrimination by 3.2 dB (or 44.5%) and 3.7 dB (or 53.1%) across temporal frequencies, with bandwidths of 3.9 and 3.1 octaves, respectively; (3) visual acuity by 3.2 dB (or 44.5%). The fellow eye also showed a small amount of improvement in contrast sensitivities and no significant change in motion perception. Control subjects who received no training demonstrated no obvious improvement in any measure. CONCLUSIONS. The results demonstrate substantial plasticity in the amblyopic visual system, and provide additional empirical support for perceptual learning as a potential treatment for amblyopia.     

Detection of chromoluminance patterns on chromoluminance pedestals I: threshold measurements

Measurement of the detection thresholds of patterns on pedestals of various kinds has the potential of providing insight into the mechanisms that mediate pattern vision. This study is concerned with chromoluminance patterns, that is, patterns that vary over space in luminance, chromaticity, or both. Contrast thresholds for 1 c/deg Gabor patterns (targets) were measured as a function of the contrast of Gabor pedestal patterns (TvC functions), where the pedestals paired with each target were modulated in a wide range of directions in color space. For most target-pedestal pairs, the TvC function decreased (facilitation) and then increased as pedestal contrast increased. The increase went above the absolute contrast threshold (masking) for all target-pedestal pairs except in cases where facilitation occurred at the upper end of the pedestal contrast range. The specific form of the TvC function varied greatly with the target and pedestal, consistent with a general model of pedestal effects proposed by Foley [Journal of the Optical Society of America A, 1994, 11(6)]. There were two sets of target-pedestal pairs for which facilitation did not occur, but masking did occur: pairs in which the target was a luminance modulation and the pedestals were individually isoluminant and pairs in which the pedestal was blue/yellow and the target was in any of our directions except blue/yellow.     

Grating and plaid masks indicate linear summation in a contrast gain pool

In human vision, the response to luminance contrast at each small region in the image is controlled by a more global process where suppressive signals are pooled over spatial frequency and orientation bands. But what rules govern summation among stimulus components within the suppressive pool? We addressed this question by extending a pedestal plus pattern mask paradigm to use a stimulus with up to three mask components: a vertical 1 c/deg pedestal, plus pattern masks made from either a grating (orientation = -45 degrees ) or a plaid (orientation = +/-45 degrees ), with component spatial frequency of 3 c/deg. The overall contrast of both types of pattern mask was fixed at 20% (i.e., plaid component contrasts were 10%). We found that both of these masks transformed conventional dipper functions (threshold vs. pedestal contrast with no pattern mask) in exactly the same way: The dipper region was raised and shifted to the right, but the dipper handles superimposed. This equivalence of the two pattern masks indicates that contrast summation between the plaid components was perfectly linear prior to the masking stage. Furthermore, the pattern masks did not drive the detecting mechanism above its detection threshold because they did not abolish facilitation by the pedestal (Foley, 1994). Therefore, the pattern masking could not be attributed to within-channel masking, suggesting that linear summation of contrast signals takes place within a suppressive contrast gain pool. We present a quantitative model of the effects and discuss the implications for neurophysiological models of the process.     

Modeling mechanisms of perceptual learning with augmented Hebbian re-weighting

Using the external noise plus training paradigm, we have consistently found that two independent mechanisms, stimulus enhancement and external noise exclusion, support perceptual learning in a range of tasks. Here, we show that re-weighting of stable early sensory representations through Hebbian learning ( [Petrov et al., 2005] and [Petrov et al., 2006]) can generate performance patterns that parallel a large range of empirical data: (1) perceptual learning reduced contrast thresholds at all levels of external noise in peripheral orientation identification ( [Dosher and Lu, 1998] and [Dosher and Lu, 1999]), (2) training with low noise exemplars transferred to performance in high noise, while training with exemplars embedded in high external noise transferred little to performance in low noise (Dosher & Lu, 2005), and (3) pre-training in high external noise only reduced subsequent learning in high external noise, whereas pre-training in zero external noise left very little additional learning in all the external noise conditions (Lu et al., 2006). In the augmented Hebbian re-weighting model (AHRM), perceptual learning strengthens or maintains the connections between the most closely tuned visual channels and a learned categorization structure, while it prunes or reduces inputs from task-irrelevant channels. Reducing the weights on irrelevant channels reduces the contributions of external noise and additive internal noise. Manifestation of stimulus enhancement or external noise exclusion depends on the initial state of internal noise and connection weights in the beginning of a learning task. Both mechanisms reflect re-weighting of stable early sensory representations.     

Effect of eccentricity on luminance-pedestal flicker thresholds

We investigated the effect that spatially coincident luminance increments (luminance pedestals) have on flicker thresholds at several eccentricities and target sizes. Luminance pedestals elevated flicker amplitude-thresholds more when stimuli were presented eccentrically, both at low (4 Hz) and high (20 Hz) temporal frequencies. Altering the size of the eccentric stimulus failed to equate central and eccentric thresholds at all pedestal amplitudes. Comparisons with flicker thresholds at various background luminances suggests that the increase in luminance-pedestal flicker thresholds peripherally is due to increased suppressive rod-cone interactions, increased effectiveness of luminous contrast on edge-sensitive flicker mechanisms, as well as increased gain in the light adaptation response.     

Orientation discrimination in foveal and extra-foveal vision: effects of stimulus bandwidth and contrast

The parameter E2 is used in many spatial scaling studies to characterize the rate at which stimulus size must increase with eccentricity to achieve foveal levels of performance in detection and discrimination tasks. We examined whether the E2 for an orientation discrimination task was dependent on the spatial frequency bandwidth of the stimulus used. Two methods were employed. In Experiments 1 and 2 stimuli were presented at a fixed high level of contrast across viewing conditions. In both experiments the E2s recovered for narrowband stimuli were larger than those recovered for broadband stimuli. In Experiment 3 we controlled for the potentially confounding effects of perceptual contrast by measuring orientation thresholds over a range of stimulus contrast levels. Only thresholds which had reached an asymptotic level, such that increases in stimulus contrast led to no further changes to thresholds, were included in the calculation of E2. We observed that E2s recovered in the latter condition were in the range of 1.29 degrees -1.83 degrees and similar for narrowband and broadband stimuli. We conclude that a failure to consider the role of perceptual contrast may result in inflated estimates of E2.     

Effect of repeated testing on contrast sensitivity

The effect of repeated testing on contrast sensitivity was examined across days and between eyes with the Nicolet CS-2000 Vision Tester. Monocular contrast sensitivity functions (CSF's) were measured with the standard technique for training days 1 to 5 on a group of 20 subjects. The group was then divided into four groups of five subjects each. A single post-training CSF was obtained for each group 1, 3, 5, or 7 days after the cessation of training. Results show that there is no measurable effect of practice on contrast sensitivity over a 5-day training period, nor is there any decrease in performance in the absence of training. CSF's were also obtained on both eyes of a separate group of subjects. We found no transfer of learning either between eyes within a single session or across multiple training sessions. We conclude that although the absence of a significant practice effect has important clinical advantages, the variability of the technique is high, allowing only robust experimental effects to be detected.     

An inherent nonlinearity in near-threshold contrast detection

We measured an essentially normal pedestal effect using stationary gaussian targets and slowly moving pedestal gratings. Since these conditions greatly reduce the information provided by the pedestal, we question whether uncertainty about the stimulus can be the main cause of the pedestal effect.     

Discriminating depth in corrugated stereo surfaces: facilitation by a pedestal is explained by removal of uncertainty

With luminance gratings, psychophysical thresholds for detecting a small increase in the contrast of a weak 'pedestal' grating are 2-3 times lower than for detection of a grating when the pedestal is absent. This is the 'dipper effect'--a reliable improvement whose interpretation remains controversial. Analogies between luminance and depth (disparity) processing have attracted interest in the existence of a 'disparity dipper'. Are thresholds for disparity modulation (corrugated surfaces), facilitated by the presence of a weak disparity-modulated pedestal? We used a 14-bit greyscale to render small disparities accurately, and measured 2AFC discrimination thresholds for disparity modulation (0.3 or 0.6 c/deg) of a random texture at various pedestal levels. In the first experiment, a clear dipper was found. Thresholds were about 2x lower with weak pedestals than without. But here the phase of modulation (0 or 180 deg) was varied from trial to trial. In a noisy signal-detection framework, this creates uncertainty that is reduced by the pedestal, which thus improves performance. When the uncertainty was eliminated by keeping phase constant within sessions, the dipper effect was weak or absent. Monte Carlo simulations showed that the influence of uncertainty could account well for the results of both experiments. A corollary is that the visual depth response to small disparities is probably linear, with no threshold-like nonlinearity.     

Preserved gain control for luminance contrast during binocular rivalry suppression

Binocular rivalry elevates contrast increment thresholds for the detection of a transient stimulus presented to the suppressed eye, while thresholds measured during dominance are identical to those during monocular viewing (e.g. [Wales, R., & Fox, R. (1970). Increment detection thresholds during binocular rivalry suppression. Perception and Psychophysics, 8, 90-94]). It is well established that contrast increment thresholds depend on reference (pedestal) contrast. With high contrasts, increment thresholds increase with pedestal contrast, reflecting a gain control with sigmoidal non-linearity. We examined how this gain control mechanism operates during binocular rivalry (i.e., with and without perception of a pedestal mask). Subjects viewed a horizontal sine-wave grating (steady pedestal) and a radial checkerboard dichoptically. When the grating achieved a pre-specified phenomenal state (dominance or suppressed), subjects initiated the transient presentation (500-ms Gaussian pulse) of a contrast increment of the same spatial frequency. The pulse appeared in either the upper or lower half of the pedestal. Subjects indicated which half of the pedestal contained the pulse. Contrast increment thresholds were measured using a staircase method with various pedestal contrasts, which yielded threshold versus contrast (TvC) functions during dominance and suppression. The measured thresholds were reliably higher during suppression, but the rising slopes of TvC functions did not differ significantly between dominance and suppression (i.e., constant upward shift of TvC function). A control experiment demonstrated that the TvC function during dominance was identical to that during non-rivalry, monocular viewing. Evidently, the contrast gain control for transient luminance increment does not require the perception of pedestal contrast.     

Optimal spatial localization is limited by contrast sensitivity.

Bisection is one of several spatial localization tasks that achieve hyperacuity performance levels. We find that optimal bisection thresholds, and hyperacuity tasks in general, are no better than might be expected from simple contrast detection and discrimination performance. The three-line bisection task can be described in terms of the test-pedestal paradigm where the test pattern is a horizontal dipole and the pedestal is a horizontal three-line pattern with equal spacing between the lines. When the dipole test is added to the center line, the line shifts up or down, depending on the test polarity. For low contrast pedestal lines at the optimal separation, the bisection threshold falls between the observer's own dipole contrast detection threshold and the bottom of the dipole contrast discrimination dipper function. At higher pedestal strengths performance degrades with a slope of about 0.5-0.7, similar to that found in contrast discrimination tasks. Therefore, bisection performance is compatible with expectations based on contrast discrimination data. At large pedestal line separations (> 10 min) bisection thresholds in min are about 1/60 the separation and relatively independent of pedestal strength. These findings are consistent with the idea that two processes are involved in limiting bisection performance; the first limit is based on contrast sensitivity of the system and the second limit to performance is based on a local sign or position tag processing. Finally, when bisection is compared with Vernier acuity and blur resolution tasks, where the test is also a dipole, bisection performance falls roughly midway, better than Vernier acuity but worse than blur resolution.     

The effect of orientation learning on contrast sensitivity

Regan and Beverley [Regan, D., & Beverley, K. I. (1985). Postadaptation orientation discrimination. Journal of the Optical Society of America A, 2(2), 147-155] previously demonstrated that adapting to an oriented visual stimulus improves sensitivity to subtle orientation differences while impairing contrast sensitivity. Here, we investigated whether practice-based improvements in orientation sensitivity would, like adaptation, impair contrast sensitivity. To the contrary, we found that contrast sensitivity actually improved significantly after observers demonstrated practice-based increases in orientation sensitivity. Therefore, while orientation sensitivity can be enhanced either by orientation-discrimination training or by adapting to visual stimuli, these two procedures have opposite effects on contrast sensitivity. This difference suggests that adaptation and perceptual learning on orientation discrimination cannot be explained sufficiently by a shared underlying cause, such as a reduction in neural activity.     

Motion perceptual learning: when only task-relevant information is learned

The classic view that perceptual learning is information selective and goal directed has been challenged by recent findings showing that subthreshold and task-irrelevant information can induce perceptual learning. This study demonstrates a limit on task-irrelevant learning as exposure to suprathreshold task-irrelevant signals failed to induce perceptual learning. In each trial, two random-dot motion stimuli were presented in a two-alternative forced-choice task. Observers either decided which of the two contained a coherent motion signal (detection task), or whether the coherent motion direction was clockwise or counterclockwise relative to a reference direction (discrimination task). Whereas the exact direction of the coherent motion signal was irrelevant to the detection task, detection of the coherent motion signal was necessary for the discrimination task. We found that the detection trainees improved only their detection but not discrimination sensitivity, whereas the discrimination trainees improved both. Therefore, the importance of task relevance was demonstrated in both detection and discrimination learning. Furthermore, both detection and discrimination training along a single pedestal direction transferred to a broad range of pedestal directions. The profile of the discrimination transfer (as a function of pedestal direction) narrowed for the discrimination trainees.     

Stimulus uncertainty in auditory perceptual learning

Stimulus uncertainty produced by variations in a target stimulus to be detected or discriminated, impedes perceptual learning under some, but not all experimental conditions. To account for those discrepancies, it has been proposed that uncertainty is detrimental to learning when the interleaved stimuli or tasks are similar to each other but not when they are sufficiently distinct, or when it obstructs the downstream search required to gain access to fine-grained sensory information, as suggested by the Reverse Hierarchy Theory (RHT). The focus of the current review is on the effects of uncertainty on the perceptual learning of speech and non-speech auditory signals. Taken together, the findings from the auditory modality suggest that in addition to the accounts already described, uncertainty may contribute to learning when categorization of stimuli to phonological or acoustic categories is involved. Therefore, it appears that the differences reported between the learning of non-speech and speech-related parameters are not an outcome of inherent differences between those two domains, but rather due to the nature of the tasks often associated with those different stimuli.     

Benefits of efficient consolidation: short training enables long-term resistance to perceptual adaptation induced by intensive testing

Intensive training or testing reduces performance on perceptual and sensorimotor tasks. Here we show, for the visual texture discrimination task, that such adaptation-related performance decrements are practically eliminated following practice with a small number of trials and sleep. Thus, short training produces consolidation of an effective memory within the visual neural network, resistant to the performance decrements that are usually induced by intensive testing. We suggest a link between perceptual adaptation and learning: resistance is achieved by sleep dependent consolidation of distributed changes in network connectivity before saturated due to over-training. This link between memory generation, perceptual adaptation and memory consolidation may have an essential role in the underlying mechanisms of perceptual and motor learning. Therefore, intensive training yielding performance decrements in other modalities, such as the sensorimotor system, may be viewed in the context of the mechanisms suggested here.     

Mechanisms of generalization in perceptual learning

Learning in many visual perceptual tasks has been shown to be specific to practiced stimuli, while new stimuli have to be learned from scratch. Here we demonstrate generalization using a novel paradigm in motion discrimination where learning has been previously shown to be specific. We trained subjects to discriminate directions of moving dots, and verified the previous results that learning does not transfer from a trained direction to a new one. However, by tracking the subjects' performance across time in the new direction, we found that their speed of learning doubled. Therefore, we found generalization in a task previously considered too difficult to generalize. We also replicated, in a second experiment, transfer following training with 'easy' stimuli, when the difference between motion directions is enlarged. In a third experiment we found a new mode of generalization: after mastering the task with an easy stimulus, subjects who have practiced briefly to discriminate the easy stimulus in a new direction generalize to a difficult stimulus in that direction. This generalization depends on both the mastering and the brief practice. The specificity of perceptual learning and the dichotomy between learning of 'easy' versus 'difficult' tasks have been assumed to involve different learning processes at different cortical areas. Here we show how to interpret these results in terms of signal detection theory. With the assumption of limited computational capacity, we obtain the observed phenomena--direct transfer and acceleration of learning--for increasing levels of task difficulty. Human perceptual learning and generalization, therefore, concur with a generic discrimination system.     

Low-level sensory plasticity during task-irrelevant perceptual learning: Evidence from conventional and double training procedures

Studies of perceptual learning have focused on aspects of learning that are related to early stages of sensory processing. However, conclusions that perceptual learning results in low-level sensory plasticity are controversial, since such learning may also be attributed to plasticity in later stages of sensory processing or in readout from sensory to decision stages, or to changes in high-level central processing. To address this controversy, we developed a novel random dot motion (RDM) stimulus to target motion cells selective to contrast polarity by ensuring the motion direction information arises only from signal dot onsets and not their offsets, and used these stimuli in the paradigm of task-irrelevant perceptual learning (TIPL). In TIPL, learning is achieved in response to a stimulus by subliminally pairing that stimulus with the targets of an unrelated training task. In this manner, we are able to probe learning for an aspect of motion processing thought to be a function of directional V1 simple cells with a learning procedure that dissociates the learned stimulus from the decision processes relevant to the training task. Our results show direction-selective learning for the designated contrast polarity that does not transfer to the opposite contrast polarity. This polarity specificity was replicated in a double training procedure in which subjects were additionally exposed to the opposite polarity. Taken together, these results suggest that TIPL for motion stimuli may occur at the stage of directional V1 simple cells. Finally, a theoretical explanation is provided to understand the data.     

Perceptual learning of bisection stimuli under roving: slow and largely specific

In perceptual learning, performance often improves within a short time if only one stimulus variant is presented, such as a line bisection stimulus with one outer-line-distance. However, performance stagnates if two bisection stimuli with two outer-line-distances are presented randomly interleaved. Recently, S. G. Kuai, J. Y. Zhang, S. A. Klein, D. M. Levi, and C. Yu, (2005) proposed that learning under roving conditions is impossible in general. Contrary to this proposition, we show here that perceptual learning with bisection stimuli under roving is possible with extensive training of 18000 trials. Despite this extensive training, the improvement of performance is still largely specific. Furthermore, this improvement of performance cannot be explained by an accommodation to stimulus uncertainty caused by roving.