Visual Working Memory Capacity Can Be Increased by Training on Distractor Filtering Efficiency

Abstract

It is generally considered that working memory (WM) capacity is limited and that WM capacity affects cognitive processes. Distractor filtering efficiency has been suggested to be an important factor in determining the visual working memory (VWM) capacity of individuals. In the present study, we investigated whether training in visual filtering efficiency (FE) could improve VWM capacity, as measured by performance on the change detection task (CDT) and changes of contralateral delay activity (CDA) (contralateral delay activity) of different conditions, and evaluated the transfer effect of visual FE training on verbal WM and fluid intelligence, as indexed by performance on the verbal WM span task and Raven's Standard Progressive Matrices (RSPM) test, respectively. Participants were divided into high- and low-capacity groups based on their performance in a CDT designed to test VWM capacity, and then the low-capacity individuals received 20 days of FE training. The training significantly improved the group's performance in the CDT, and their CDA models of different conditions became more similar with high capacity group, and the effect generalized to improve verbal WM span. These gains were maintained at a 3-month follow-up test. Participants' RSPM scores were not changed by the training. These findings support the notion that WM capacity is determined, at least in part, by distractor FE and can be enhanced through training.

Keywords: CDA; capacity; filtering efficiency; training; visual working memory.

FIGURE 1

General training procedure.

FIGURE 2

Visual working memory (VWM) task procedure. A filtering condition example is shown.

FIGURE 3

Training paradigms: remember the color on the cue side (A). Remember the shape on the cue side (B). Remember the location of items on the cue side (C). Remember the shape of the red items (D). Remember the location of the red items (E). Remember the color of triangles (G).

FIGURE 4

Mean accuracy (A) and Cowan K-values (B) for high and low capacity groups at the pre-training, post-training, and follow-up assessment. Asterisks indicate statistical differences at ^∗p < 0.05 or ^∗∗p < 0.01.

FIGURE 5

Inverse correlation between individuals’ memory capacity scores and US.

FIGURE 6

Verbal WM performance. Group VeWMST scores at pre-training, post-training, and follow-up assessments. Asterisks indicate statistical differences at ^∗∗p < 0.01.

FIGURE 7

Contralateral delay activity (CDA) waveforms and CDA amplitude means. Grand average ERP difference waves for the low capacity group at pretest (A), posttest (C), and follow-up (D). (B) Grand average ERP difference waves of high capacity individuals at pretest. (E) CDA amplitudes from each of the three conditions for each group at pre-training, post-training, and follow-up. And as means approach significant. Asterisks indicate statistical differences at ^∗p < 0.05 or ^∗∗p < 0.01.

FIGURE 8

Correlation between an individual’s memory capacity and filtering efficiency.