Spontaneous and Training‐Induced Visual Learning in Cortical Blindness: Characteristics and Neural Substrates

Visual learning has been intensively studied in higher mammals, both during development and in adulthood. What is less clear is the extent and properties such plasticity may acquire following permanent damage to the adult visual system. Answering this question is important. Aside from improving our understanding of visual processing in the absence of an intact visual circuitry, such knowledge is essential for the development of effective therapies to rehabilitate the increasing number of people who suffer the functional consequences of damage at different levels of their visual cortical hierarchy. This review summarizes the known characteristics of visual learning after adult visual cortex damage and begins to dissect some of the neural correlates of this process.     

Shared Mechanisms of Perceptual Learning and Decision Making

Perceptual decisions require the brain to weigh noisy evidence from sensory neurons to form categorical judgments that guide behavior. Here we review behavioral and neurophysiological findings suggesting that at least some forms of perceptual learning do not appear to affect the response properties of neurons that represent the sensory evidence. Instead, improved perceptual performance results from changes in how the sensory evidence is selected and weighed to form the decision. We discuss the implications of this idea for possible sites and mechanisms of training‐induced improvements in perceptual processing in the brain.     

Perceptual learning in amnesia

Evidence from experiments on perceptual learning, accumulated during the last few years, increasingly indicates that the relative 'front end' parts of the visual system are more plastic even in adults than was previously expected. Hence, it might be possible that perceptual learning is similar in several respects to procedural learning and may be achieved even without (declarative) memory traces present. Results on six patients suffering from global amnesia due to damage to hippocampal-diencephalic systems demonstrate, for the first time, that at least some amnesic patients are able to significantly improve performance in a visual hyperacuity task as a result of training, showing improvement as good as the observers in the control group. This result corroborates the notion of a relatively 'front end' location of at least some forms of perceptual learning.     

Stimulus specific adaptation in excited but not in inhibited cells in inferotemporal cortex of Macaque

Many cells in inferotemporal cortex respond more actively to a novel presentation than to a subsequent re-presentation of the same image, exhibiting stimulus specific adaptation (SSA). Previously, analysis of this adaptation was limited to visually excited cells, excluding visually inhibited cells. In the present experiment we studied 654 cells in four macaques performing visual tasks. Strong SSA (P < 0.0001) was observed in those cells which were excited by visual stimuli. This adaptation was also seen in the subset of such cells which, though excited by visual stimuli, failed to show visual specificity in their responses. Interestingly, no SSA (P > 0.1) was observed in the group of cells inhibited by visual stimuli. Furthermore, most inhibited cells failed to show visual specificity. This lack of visual specificity and SSA suggest that the visually inhibited cells have a limited role in the detailed information processing of visual perception and memory activated by the tasks used in the present experiments.     

Perceptual Learning During Action Video Game Playing

Action video games have been shown to enhance behavioral performance on a wide variety of perceptual tasks, from those that require effective allocation of attentional resources across the visual scene, to those that demand the successful identification of fleetingly presented stimuli. Importantly, these effects have not only been shown in expert action video game players, but a causative link has been established between action video game play and enhanced processing through training studies. Although an account based solely on attention fails to capture the variety of enhancements observed after action game playing, a number of models of perceptual learning are consistent with the observed results, with behavioral modeling favoring the hypothesis that avid video game players are better able to form templates for, or extract the relevant statistics of, the task at hand. This may suggest that the neural site of learning is in areas where information is integrated and actions are selected; yet changes in low‐level sensory areas cannot be ruled out.     

Vertical meridian representation on the prelunate gyrus in area V4 of macaque

The representation of the lower quadrant in area V4 is presently thought to extend along the prelunate gyrus from a foveal representation laterally all the way to the dorsal end of the superior temporal sulcus. However, several studies suggest the possibility of a more complex organization. To see if the visuotopic organization on the crown of the gyrus was relatively homogeneous or instead contained inhomogeneities indicative of more complex organization, we recorded from a grid of points over the prelunate gyrus. Receptive-field size and scatter in the region are large, making it difficult to infer topography from simple inspection of receptive-field sequences. We developed an averaging procedure using data from all recording sites to detect an inhomogeneity in topography with respect to the vertical meridian. With this procedure, we found a vertical meridian representation just medial to the medial end of the lateral sulcus. We also found a significant difference in the incidence of orientation sensitivity on either side of the meridian representation. The results show that the crown of the prelunate gyrus cannot be described as a single homogeneous region, but instead contains at least two different sub-regions adjoining along a shared representation of the vertical meridian.     

Perceptual Learning beyond Perception: Mnemonic Representation in Early Visual Cortex and Intraparietal Sulcus

The ability to discriminate between stimuli relies on a chain of neural operations associated with perception, memory and decision-making. Accumulating studies show learning-dependent plasticity in perception or decision-making, yet whether perceptual learning modifies mnemonic processing remains unclear. Here, we trained human participants of both sexes in an orientation discrimination task, while using functional magnetic resonance imaging (fMRI) and transcranial magnetic stimulation (TMS) to separately examine training-induced changes in working memory (WM) representation. fMRI decoding revealed orientation-specific neural patterns during the delay period in primary visual cortex (V1) before, but not after, training, whereas neurodisruption of V1 during the delay period led to behavioral deficits in both phases. In contrast, both fMRI decoding and disruptive effect of TMS showed that intraparietal sulcus (IPS) represented WM content after, but not before, training. These results suggest that training does not affect the necessity of sensory area in representing WM information, consistent with the sensory recruitment hypothesis in WM, but likely alters the coding format of the stored stimulus in this region. On the other hand, training can render WM content to be maintained in higher-order parietal areas, complementing sensory area to support more robust maintenance of information.SIGNIFICANCE STATEMENT There has been accumulating progresses regarding experience-dependent plasticity in perception or decision-making, yet how perceptual experience moulds mnemonic processing of visual information remains less explored. Here, we provide novel findings that learning-dependent improvement of discriminability accompanies altered WM representation at different cortical levels. Critically, we suggest a role of training in modulating cortical locus of WM representation, providing a plausible explanation to reconcile the discrepant findings between human and animal studies regarding the recruitment of sensory or higher-order areas in WM.     

Oscillatory Neuronal Responses in the Visual Cortex of the Awake Macaque Monkey

An important step in early visual processing is the segmentation of scenes. Features constituting individual objects have to be grouped together and segregated from those of other figures or the background. It has been proposed that this grouping could be achieved by synchronizing the fine temporal structure of responses from neurons excited by an individual figure. In the cat visual cortex evidence has been obtained that responses of feature-selective neurons have a distinctive oscillatory structure and can synchronize both within and across cortical areas, the synchronization depending on stimulus configuration. Here we investigate the generality of oscillatory responses and their synchronization and specifically whether these phenomena occur in extrastriate areas of the visual cortex of the awake behaving primate. We find in the caudal superior temporal sulcus of the macaque monkey ( Macaca fascicularis ) that adjacent neurons can synchronize their responses, in which case their discharges exhibit an oscillatory temporal structure. During such periods of local synchrony spatially separated cell groups can also synchronize their responses if activated with a single stimulus. These findings resemble those described previously for the cat visual cortex, except that in the awake monkey the oscillatory episodes tend to be of shorter duration and exhibit more variability of oscillation frequency.     

Neuronal Generators of Visual Evoked Potentials in Humans: Visual Processing in the Human Cortex

Summary: Purpose : We wished to define the localization of cortical generators of visual (pattern) evoked potentials (VEP) and the temporal sequence of activation in the occipital region.
Methods : In 4 candidates for epilepsy surgery, a large array of subdural electrodes was placed over occipital areas. Checkerboard pattern reversal stimuli were generated and the epileptogenic focus was localized and functionally mapped. Magnetic resonance imaging did not show any occipital lesions in any of the 4 patients.
Results : The area first activated was the lingual gyrus in the mesial occipital lobe (negative potential peaks at ∼70 ms), followed by an area superior to the calcarine fissure (negative peaks at ∼80 ms). Later (starting at ∼90 ms), there were positive potentials over the occipital pole and lingual gyrus, followed by potentials at the lateral occipital lobe.
Conclusions : These data support the idea that VEP are generated in the mesial and lateral occipital cortex by different circumscribed neuronal generators with different latencies of activation. The scalp-recorded N1 and P1 potential peaks most likely derive from the progressive activation of neuronal masses in different regions of the occipital lobe.     

Visual evoked potentials in acute occipital blindness

Summary Pattern-evoked and flash-evoked responses were recorded from four patients with acute blindness due to occipital ischaemia. No responses could be obtained with pattern stimulation. Flash stimulation, however, yielded well-reproducible potentials in all four cases in spite of practically complete visual loss. In one patient vision recovered completely while the three others showed poor or minimal recovery in restricted visual field areas only. It is concluded that the flash method is not appropriate for differentiation of occipital blindness from psychogenic visual disorders. Furthermore, preserved flashevoked potentials in the acute stage of occipital blindness cannot be taken as a reliable prognostic sign for visual recovery, as has been postulated by others. A possible extrastriate origin of these flash responses is discussed in view of recent theories on a second visual system.Presented in part at the conference of the German Neurological Society in Munich (15–17 October 1981)     

Reduced Learning of Sound Categories in Dyslexia Is Associated with Reduced Regularity-Induced Auditory Cortex Adaptation

A main characteristic of dyslexia is poor use of sound categories. We now studied within-session learning of new sound categories in dyslexia, behaviorally and neurally, using fMRI. Human participants (males and females) with and without dyslexia were asked to discriminate which of two serially-presented tones had a higher pitch. The task was administered in two protocols, with and without a repeated reference frequency. The reference condition introduces regularity, and enhances frequency sensitivity in typically developing (TD) individuals. Enhanced sensitivity facilitates the formation of “high” and “low” pitch categories above and below this reference, respectively. We found that in TDs, learning was paralleled by a gradual decrease in activation of the primary auditory cortex (PAC), and reduced activation of the superior temporal gyrus (STG) and left posterior parietal cortex (PPC), which are important for using sensory history. No such sensitivity was found among individuals with dyslexia (IDDs). Rather, IDDs showed reduced behavioral learning of stimulus regularities and no regularity-associated adaptation in the auditory cortex or in higher-level regions. We propose that IDDs'' reduced cortical adaptation, associated with reduced behavioral learning of sound regularities, underlies their impoverished use of stimulus history, and consequently impedes their formation of rich sound categories.SIGNIFICANCE STATEMENT Reading difficulties in dyslexia are often attributed to poor use of phonological categories. To test whether poor category use could result from poor learning of new sound categories in general, we administered an auditory discrimination task that examined the learning of new pitch categories above and below a repeated reference sound. Individuals with dyslexia (IDDs) learned categories slower than typically developing (TD) individuals. TD individuals showed adaptation to the repeated sounds that paralleled the category learning in their primary auditory cortex (PAC) and other higher-level regions. In dyslexia, no brain region showed such adaptation. We suggest that poor learning of sound statistics in sensory regions may underlie the poor representations of both speech and nonspeech categories in dyslexia.     

To the Trained Eye: Perceptual Expertise Alters Visual Processing

Perceptual expertise refers to learning that is specific to a domain, that transfers to new items within the trained domain, and that leads to automatic processing in the sense that expertise effects can be measured across a variety of tasks. It can be argued that most of us possess some degree of perceptual expertise in a least one, if not several domains, thereby giving the study of perceptual expertise broad application. Some object categories may in fact be objects of perceptual expertise to the majority of people: Faces appear to be one such example. Thus, the use of face stimuli, or the comparison of face and object perception, can be a powerful way to ask whether a given process is influenced by perceptual expertise. Here, we emphasize one characteristic way that face processing appears to differ from nonface processing: that is, the degree to which they recruit a “holistic” rather than a “featural” perceptual strategy. This review brings evidence that expertise influences perceptual processing together with recent findings that the capacity of visual short‐term memory is greater in perceptual experts and explores the relationship between the two.     

Perceptual deficits in amnesia: challenging the medial temporal lobe 'mnemonic' view

Recent animal studies suggest that the medial temporal lobe (MTL), which is thought to subserve memory exclusively, may support non-mnemonic perceptual processes, with the hippocampus and perirhinal cortex contributing to spatial and object perception, respectively. There is, however, no support for this view in humans, with human MTL lesions causing prominent memory deficits in the context of apparently normal perception. We assessed visual discrimination in amnesic cases to reveal that while selective hippocampal damaged patients could discriminate faces, objects, abstract art and colour, they were significantly poorer in discriminating spatial scenes. By contrast, patients with MTL damage, including perirhinal cortex, were significantly impaired in discriminating scenes, faces, and to a lesser extent objects, with relatively intact discrimination of art and colour. These novel observations imply that the human MTL subserves both perceptual and mnemonic functions, with the hippocampus and perirhinal cortex playing distinct roles in spatial and object discrimination, respectively.     

Development of neuronal responses in cat posteromedial lateral suprasylvian visual cortex

Exchange assays have often been used to quantitate steroid receptors when endogenous ligands are present; however, there are no reports of their successful application to activated glucocorticoid-Type II receptor complexes. In addition to investigating the reasons for this failure, the present study also examined the effects of progesterone on glucocorticoid dissociation from, and reassociation with unactivated and activated Type II receptors. Molybdate-stabilized brain cytosol from adrenal-ovariectomized mice was incubated with [3H]dexamethasone ( +/- [1H]DEX) for 40 h at 0 degree C. Afterwards free steroid was removed on Sephadex G-25 columns in the presence (unactivated receptors) or absence (activated receptors) of molybdate. Activation, as measured by DNA-cellulose binding, was achieved by incubating molybdate-free cytosol at 22 degrees C for 20 min followed by G-25 filtration in the presence of molybdate. The rates of dissociation and reassociation were then measured by incubating cytosol with [1H]triamcinolone acetonide (TA) or [3H]TA ( +/- [1H]TA) at 12 degrees C. An exchange assay was also employed in which cytosol was incubated first with [1H]DEX for 40 h at 0 degree C followed by bound-free steroid separations and 12 degrees C incubations with [3H]TA ( +/- [1H]TA). Both approaches revealed that even though activation reduced the rate of DEX dissociation from Type II receptors by 40%, it eliminated the ability of the newly unoccupied receptors to rebind glucocorticoid. Adding [1H]progesterone to occupied receptor preparations increased dissociation rate constants by nearly 3-fold, for both unactivated and activated Type II receptors. Since [1H]TA failed to prevent this effect, progesterone appears to act at an allosteric site(s) which cannot be occupied by glucocorticoids. Exchange assays revealed that progesterone-facilitated dissociation increased the rate of glucocorticoid rebinding to unactivated, but not activated Type II receptors. These results suggest that spontaneous and progesterone-facilitated termination of glucocorticoid genomic actions could be mediated by steroid dissociation since unoccupied activated Type II receptors do not rebind agonist steroid.     

Interactions between attention and perceptual grouping in human visual cortex

Freeman et al. demonstrated that detection sensitivity for a low contrast Gabor stimulus improved in the presence of flanking, collinearly oriented grating stimuli, but only when observers attended to them. By recording visual event-related potentials (ERPs) elicited by a Gabor stimulus, we investigated whether this contextual cueing effect involves changes in the short-latency afferent visual signal from V1 that have a stimulus onset latency between 60 and 80 ms and/or longer-latency changes from visual cortex. Under dual-task conditions, the subjects performed contrast discrimination for a central Gabor and an orientation judgment for a pre-specified subset of the flanking Gabors. On random trials, the central Gabor could be collinearly or orthogonally oriented with respect to the attended flankers. Subjects showed improvements in discriminating the contrast of the central grating when it was oriented collinearly with the attended flankers. The ERP difference between attending to collinear versus orthogonal flankers manifested as a positive polarity response at occipital electrodes with a latency of 180-250 ms after stimulus onset. No shorter-latency contextual cueing differences were observed in the ERPs. The ERP latency profile of the contextual cueing effect argues against the hypothesis that short-latency afferent activity from V1 is the stage of processing at which attention can influence neuronal lateral interactions. However, the scalp voltage distribution of the longer-latency contextual cueing effect is similar to the one generated by the early phasic stimulus onset activity from V1. These findings leave open the possibility that V1 is involved in the attentional modulation of lateral interactions but that this has a longer time course, likely being mediated by re-afferent inputs from later stages of the visual pathway.     

Perceptual discrimination in fear generalization: Mechanistic and clinical implications

For almost a century, Pavlovian conditioning is the imperative experimental paradigm to investigate the development and generalization of fear. However, despite the rich research tradition, the conceptualization of fear generalization has remained somewhat ambiguous. In this selective review, we focus explicitly on some challenges with the current operationalization of fear generalization and their impact on the ability to make inferences on its clinical potential and underlying processes. The main conclusion is that, despite the strong evidence that learning influences perception, current research has largely neglected the role of perceptual discriminability and its plasticity in fear generalization. We propose an alternative operationalization of generalization, where the essence is that Pavlovian conditioning itself influences the breadth of fear generalization via learning-related changes in perceptual discriminability. Hence a conceptualization of fear generalization is incomplete without an in-depth analysis of processes of perceptual discriminability. Furthermore, this highlights perceptual learning and discriminability as important future targets for pre-clinical and clinical research.     

Dissecting the Roles of Supervised and Unsupervised Learning in Perceptual Discrimination Judgments

Our ability to compare sensory stimuli is a fundamental cognitive function, which is known to be affected by two biases: choice bias, which reflects a preference for a given response, and contraction bias, which reflects a tendency to perceive stimuli as similar to previous ones. To test whether both reflect supervised processes, we designed feedback protocols aimed to modify them and tested them in human participants. Choice bias was readily modifiable. However, contraction bias was not. To compare these results to those predicted from an optimal supervised process, we studied a noise-matched optimal linear discriminator (Perceptron). In this model, both biases were substantially modified, indicating that the “resilience” of contraction bias to feedback does not maximize performance. These results suggest that perceptual discrimination is a hierarchical, two-stage process. In the first, stimulus statistics are learned and integrated with representations in an unsupervised process that is impenetrable to external feedback. In the second, a binary judgment, learned in a supervised way, is applied to the combined percept.SIGNIFICANCE STATEMENT The seemingly effortless process of inferring physical reality from the sensory input is highly influenced by previous knowledge, leading to perceptual biases. Two common ones are contraction bias (the tendency to perceive stimuli as similar to previous ones) and choice bias (the tendency to prefer a specific response). Combining human psychophysical experiments with computational modeling we show that they reflect two different learning processes. Contraction bias reflects unsupervised learning of stimuli statistics, whereas choice bias results from supervised or reinforcement learning. This dissociation reveals a hierarchical, two-stage process. The first, where stimuli statistics are learned and integrated with representations, is unsupervised. The second, where a binary judgment is applied to the combined percept, is learned in a supervised way.     

Altered Heterosynaptic Plasticity Impairs Visual Discrimination Learning in Adenosine A1 Receptor Knock-Out Mice

Theoretical and modeling studies demonstrate that heterosynaptic plasticity—changes at synapses inactive during induction—facilitates fine-grained discriminative learning in Hebbian-type systems, and helps to achieve a robust ability for repetitive learning. A dearth of tools for selective manipulation has hindered experimental analysis of the proposed role of heterosynaptic plasticity in behavior. Here we circumvent this obstacle by testing specific predictions about the behavioral consequences of the impairment of heterosynaptic plasticity by experimental manipulations to adenosine A1 receptors (A1Rs). Our prior work demonstrated that the blockade of adenosine A1 receptors impairs heterosynaptic plasticity in brain slices and, when implemented in computer models, selectively impairs repetitive learning on sequential tasks. Based on this work, we predict that A1R knock-out (KO) mice will express (1) impairment of heterosynaptic plasticity and (2) behavioral deficits in learning on sequential tasks. Using electrophysiological experiments in slices and behavioral testing of animals of both sexes, we show that, compared with wild-type controls, A1R KO mice have impaired synaptic plasticity in visual cortex neurons, coupled with significant deficits in visual discrimination learning. Deficits in A1R knockouts were seen specifically during relearning, becoming progressively more apparent with learning on sequential visual discrimination tasks of increasing complexity. These behavioral results confirm our model predictions and provide the first experimental evidence for a proposed role of heterosynaptic plasticity in organism-level learning. Moreover, these results identify heterosynaptic plasticity as a new potential target for interventions that may help to enhance new learning on a background of existing memories.SIGNIFICANCE STATEMENT Understanding how interacting forms of synaptic plasticity mediate learning is fundamental for neuroscience. Theory and modeling revealed that, in addition to Hebbian-type associative plasticity, heterosynaptic changes at synapses that were not active during induction are necessary for stable system operation and fine-grained discrimination learning. However, lacking tools for selective manipulation prevented behavioral analysis of heterosynaptic plasticity. Here we circumvent this barrier: from our prior experimental and computational work we predict differential behavioral consequences of the impairment of Hebbian-type versus heterosynaptic plasticity. We show that, in adenosine A1 receptor knock-out mice, impaired synaptic plasticity in visual cortex neurons is coupled with specific deficits in learning sequential, increasingly complex visual discrimination tasks. This provides the first evidence linking heterosynaptic plasticity to organism-level learning.     

A psychophysical study of visual extinction: ipsilesional distractor interference with contralesional orientation thresholds in visual hemineglect patients

Visual extinction was investigated in left (n=15) and right (n=25) brain-damaged patients with or without visual neglect, and in normal control subjects (n=14), using a psychophysical paradigm. Orientation discrimination thresholds were determined for both left and right hemifield gratings presented either in isolation or simultaneously with a contralateral distractor grating. To minimize the influence of possible sensory-perceptual deficits, the luminances of both target and distractor gratings were chosen to be 20 times the luminances necessary to discriminate between horizontal and vertical grating orientations. The location of the target grating was always cued, making the distractor grating task irrelevant. Even after equalizing the visibility of left and right hemifield stimuli, neglect patients still displayed an increased interference effect from an ipsilesional distractor (and no interference from a contralesional distractor). Left or right brain-damaged controls did not show this asymmetric interference of irrelevant distractors, even the patients who demonstrated extinction on standard extinction testing. This suggests that visual extinction is a critical component of the visual neglect syndrome and that it involves an attentional deficit.     

Evaluating Visual Cues Modulates Their Representation in Mouse Visual and Cingulate Cortex

Choosing an action in response to visual cues relies on cognitive processes, such as perception, evaluation, and prediction, which can modulate visual representations even at early processing stages. In the mouse, it is challenging to isolate cognitive modulations of sensory signals because concurrent overt behavior patterns, such as locomotion, can also have brainwide influences. To address this challenge, we designed a task, in which head-fixed mice had to evaluate one of two visual cues. While their global shape signaled the opportunity to earn reward, the cues provided equivalent local stimulation to receptive fields of neurons in primary visual (V1) and anterior cingulate cortex (ACC). We found that mice evaluated these cues within few hundred milliseconds. During this period, ∼30% of V1 neurons became cue-selective, with preferences for either cue being balanced across the recorded population. This selectivity emerged in response to the behavioral demands because the same neurons could not discriminate the cues in sensory control measurements. In ACC, cue evaluation affected a similar fraction of neurons; emerging selectivity, however, was stronger than in V1, and preferences in the recorded population were biased toward the cue promising reward. Such a biased selectivity regime might allow the mouse to infer the promise of reward simply by the overall level of activity. Together, these experiments isolate the impact of task demands on neural responses in mouse cerebral cortex, and document distinct neural signatures of cue evaluation in V1 and ACC.SIGNIFICANCE STATEMENT Performing a cognitive task, such as evaluating visual cues, not only recruits frontal and parietal brain regions, but also modulates sensory processing stages. We trained mice to evaluate two visual cues, and show that, during this task, ∼30% of neurons recorded in V1 became selective for either cue, although they provided equivalent visual stimulation. We also show that, during cue evaluation, mice frequently move their eyes, even under head fixation, and that ignoring systematic differences in eye position can substantially obscure the modulations seen in V1 neurons. Finally, we document that modulations are stronger in ACC, and biased toward the reward-predicting cue, suggesting a transition in the neural representation of task-relevant information across processing stages in mouse cerebral cortex.