Long-Term Visual Memory Has a Role in Suppressing Learning
Key findings
- Training people to search for common visual features in the environment led to a persistent improvement in performance over consecutive days, but also suppressed the subsequent ability to learn similar visual features
- That suppression was reversed if an interference task was used to prevent the memory from consolidating
- These findings support the concept of a neural mechanism that allows salient sensory patterns to persist in memory for prolonged periods, but that also functions to prevent false-positive detection by suppressing new learning
Subscribe to the latest updates from Neuroscience Advances in Motion
Effective visual search requires an interplay between perception, attention, executive function and oculomotor systems, but the role of memory in the task is unclear. However, previous research has shown that repeated search for specific visual features can improve with training, which suggests visual search is influenced by some form of long-term memory.
According to Gabriel N. Friedman, MD, neurosurgery resident, Ziv M. Williams, MD, neurosurgeon at Massachusetts General Hospital, and colleagues, long-term memory is a fundamental component of learning that facilitates the swath of skills and behaviors critical to human survival. At its core, they say, learning allows improved performance over time in activities such as playing the piano, but it's unknown whether learning and long-term memory underlie any basic sensory processes.
Their research, published in Frontiers in Psychology, sheds new light on the subject, supporting the existence of a neural mechanism that enables sensory patterns to persist in memory, but subsequently prevents mistakes in visual pattern identification by suppressing new learning.
The team conducted a series of three experiments designed to build on scientific understanding of the long-term memory process. Each experiment required participants to distinguish target items from similar distractor items presented on a screen. Study subjects were nine men and six women, ages 19 to 31, each of whom took part in only one experiment.
In each experiment, participants were asked to distinguish a summer clothing item from an array of winter clothing items and remember its location. After the items are reviewed on screen, they are erased and replaced with a multicolored visual mask. Then, the participants were given numbers at the locations where items had previously been presented. and asked in which location, if any, they had observed a piece of summer clothing. Their performance was measured according to accuracy, not reaction time.
[Insert figure 1 image]
Experiment 1
Experiment one assessed baseline learning performance. Participants searched for either upper- or lower-body summer clothing in randomly interspersed trials.
The results showed that participants improved over the course of training and were able to learn lower-body and upper-body items simultaneously.
Experiment 2
Experiment two evaluated whether searching for items within a particular category could lead to sustained improvements in performance after 24 hours and whether this improvement was selective to the items being searched.
On day one, the participants searched for upper-body summer clothing among winter clothing. On day two they had to identify either upper-body or lower-body summer clothing, again out of an array of winter clothing.
The researchers were surprised to observe that the training on day one (upper-body clothing) suppressed the participants' subsequent ability on day two, to learn similar visual features (lower-body clothing). This lack of improvement was not due to difficulty in searching for upper- and lower-body items simultaneously.
Experiment 3
Experiment three was identical to experiment two, except that after the initial training on day one, participants completed an "interference task" of searching unrelated items (cats vs. dogs). The purpose was to test whether the interference task would influence the learning-suppression effect noted in experiment two and confirm that this effect is memory-selective.
Introducing the interference task after training on upper-body clothing did not lead to reduced performance on upper-body clothing the following day. Rather, the interference test prevented the suppression of performance improvement for lower-body clothing (i.e., it prevented learning suppression).
Interpretation
The findings suggest a neural mechanism that enables visual patterns to persist in memory for prolonged periods of time, but that suppresses the ability to learn similar visual patterns once consolidation occurs.
The benefit of this system, the researchers propose, might be to enhance specificity to trained visual features within the environment. By suppressing the ability to learn similar but distinct visual features, the system would also reduce the possibility of false-positive visual detection.
view original journal article Subscription may be required
Refer a patient to the Department of Neurosurgery
Learn more about Mass General Neuroscience