Visual Perceptual Learning: Definition, Mechanisms, and Neural Basis

Visual perceptual learning (VPL) is defined as a long-term improvement in performance on a visual task. In recent years, the idea that conscious effort is necessary for VPL to occur has been challenged by research suggesting the involvement of more implicit processing mechanisms, such as reinforcement-driven processing and consolidation.

To fully understand VPL, it's important to consider the underlying visual perceptual skills that enable us to interpret the world around us.How to Improve Your Visual Perception SkillsVisual Perceptual skills involve the ability to organize and interpret the information that is seen and give it meaning. Our eyes send large amounts of information to our brains to process every single second. If our eyes are sending us the proper information in a way that makes sense, the brain can then process it, thus allowing us to form thoughts, make decisions, and create action.

The seven core visual perceptual skills are:

1. Visual Memory - the visual skill that allows us to record, store and retrieve information. It allows us to learn and later recall what is learned.
2. Visual Sequential Memory - similar to visual memory in that it allows us to store and retrieve information when necessary or useful. However sequential memory helps us remember and recognize people, places we have been, and series of events, equations, and procedures.
3. Visual Form Constancy - the visual skill that allows us to distinguish one object from another similar object. Being able to tell the difference between the letter “b” and “d” or “3” and “8”. Though the forms are similar in shape, they are very different in meaning. The ability to see and distinguish these differences is form constancy.
4. Visual Figure Ground - the visual skill that allows us to distinguish, segregate, isolate or find an object or stimuli in varying environments. This can include faces, figures, objects, landscapes, and letters or numbers. Properly processing our visual figure ground helps to organize the information we see in our environment.
5. Visual Spatial Relations - the visual skill that allows us to process the visual environment around us and the location of objects in respect to ourselves.
6. Visual Closure - the visual skill that allows us to detect, differentiate, select, draw conclusions and understand information when we are only given certain pieces of information, rather than the entire account, story or explanation.
7. Visual Discrimination - each of the above six skills require some degree of visual discrimination. Visual Discrimination is the ability to identify detail, seeing items likes and differences in shape, color, position and orientation.

Theoretical Framework of Visual Perceptual Learning

According to one model, visual perceptual learning (VPL) of a presented visual feature occurs when a bottom-up signal from the feature is boosted by attention or by reinforcement signals. Attention enhances task-relevant signals and inhibits task-irrelevant signals, leading to task-relevant VPL.

Figure 1: Model of visual perceptual learning.

Common Tasks Used in VPL Studies

Several tasks are commonly used in VPL research to examine different aspects of visual learning:

• Vernier acuity task: A configuration of two or three vertical lines (or dots) is presented. The subject is asked to indicate whether the lines (or the dots) are aligned.
• RSVP (rapid serial visual presentation) task: The subject is asked to identify two target items (such as white letters) within a sequence of non-target items (such as black letters) at the centre of the display. The background display consists of coherent motion (dots moving in the same direction at the same speed) and random motion (dots moving in random directions with random speed).
• Texture-discrimination task: The subject is first asked to respond according to whether a ‘T’ or an ‘L’ is presented in the centre of the display to ensure fixation at the centre, and then to indicate whether the orientation of the target is vertical or horizontal. VPL of the target orientation is examined.

Figure 2: Examples of tasks used in VPL studies.

Neural Correlates of Visual Perceptual Learning

VPL is associated with changes in various brain regions involved in visual processing and decision-making. Some experiments have indicated that training on a visual task changes visual representations in the early stages of visual signal processing, such as the tuning properties and activity of the primary visual cortex (V1) region that retinotopically corresponds to the location of the trained stimulus in the visual field. Others have instead suggested that training alters the weight of connections between the visual cortex and regions of the brain involved in decision making, or within the decision-making regions themselves.

Figure 3: Brain regions thought to be altered by visual perceptual learning (VPL).

In VPL of motion or a feature carrying spatial information, the weight changes may predominantly occur between areas in the higher visual cortex, such as the middle temporal area (MT) and the lateral intraparietal area (LIP), which is thought to be involved in visual decision-making processes.

A study using a motion direction discrimination task with training over 10 days revealed behavioral improvements and changes in brain activity. Participants were instructed to report whether the direction of the motion stimulus was clockwise or counterclockwise relative to the orientation of a red cross. The LBA (Linear Ballistic Accumulator) model was used to analyze the decision-making process, assuming one accumulator for each decision option.

Figure 1: Schematic illustrations of the experimental design and LBA model.

The study found that training improved behavioral performance, as indicated by increased accuracy and reduced response times. The drift rate, which represents the speed of evidence accumulation, was also found to increase with training.

Figure 2: Behavioral and modeling results.

The fMRI results showed that the drift rate was correlated with activity in several brain areas, including the ventral premotor cortex (PMv), supplementary eye field (SEF), and the fronto-parietal network (FPN), which includes the frontal eye field (FEF) and intraparietal cortex (IPS). These areas showed increased percentage signal changes with training.

Figure 3: Drift rate-correlated independent components and the percentage signal changes of the fMRI signals in these areas.

However, percentage signal changes in the motion selective sensory areas (V3A and MT+) were not significantly altered with training.

Figure 4: Percentage signal changes of the fMRI signals from the motion selective sensory areas (V3A and MT+).

Dynamic causal modeling (DCM) revealed that training modulated the connection from V3A to PMv and from IPS to FEF. These findings suggest that VPL involves changes in both sensory and decision-making areas of the brain.

Figure 5: DCM results.