Affective evidence that inhibition is involved in separating accessory representations from active representations in visual working memory

ABSTRACT

The multiple state theory of working memory suggests that representations are divided into two states: focused-on active representations and accessory memories held for later use. Here we tested two competing hypotheses regarding the neurocognitive mechanisms responsible for this separation: (1) that accessory memories undergo inhibition or (2) that accessory memories are amplified less than active representations. We explored whether accessory memories undergo affective devaluation, a known index of the involvement of inhibition in a visual task. On each trial participants memorized four items, were cued to focus on one, and then completed a visual search or an affective evaluation task. While search distractors matching the colour of an active item slowed search, those matching an accessory memory did not, replicating previous findings that only active items guide search. Also, accessory items were affectively devalued compared to baseline and active items, supporting the hypothesis that accessory memories undergo inhibition.

KEYWORDS:

• Working memory
• retro-cue
• affective evaluation
• attentional amplification
• inhibition
• devaluation-by-inhibition
• accessory memory

Acknowledgement

Our thanks to Michaela Comrie for her help with data collection and to Gavin Petrie for his help with earlier investigations of these concepts.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Mark J. Fenske http://orcid.org/0000-0003-4338-7754

Notes

1 As confirmation that our separation of groups in terms of pattern of search times was effective, a 2 (Group: Capture vs. No Capture) × 2 (SingletonType: Accessory-matching vs. Active-matching) mixed-factors analysis of variance revealed a significant SingletonType × Group interaction, F(1,25) = 23.34, p < .001, . Indeed, Figure 4 shows that search times for participants in the Capture Group were slower on trials with an active-matching singleton compared to trials with an accessory-matching singleton, t(14) = 3.86, p = .002, whereas participants in the No-Capture Group had search times that were slower on trials with an accessory-matching singleton compared to trials with an active-matching singleton, t(11) = 4.02, p = .002.

Additional information

Funding

This research was supported by the Natural Science and Engineering Research Council of Canada, the Canada Foundation for Innovation, and the Ontario Ministry of Research and Innovation.