EEG and pupillometric signatures of working memory overload

Abstract

Understanding the physiological correlates of cognitive overload has implications for gauging the limits of human cognition, for developing novel methods to define cognitive overload, and for mitigating the negative outcomes associated with overload. Most previous psychophysiological studies manipulated verbal working memory load in a narrow range (an average load of 5 items). It is unclear, however. how the nervous system responds to working memory load exceeding typical capacity limits. The objective of the current study was to characterize the central and autonomic nervous system changes associated with memory overload, by means of combined recording of EEG and pupillometry. Eighty-six participants were presented with a digit span task involving the serial auditory presentation of items. Each trial consisted of sequences of either 5, 9, or 13 digits, each separated by 2 seconds. Both theta activity and pupil size, after the initial rise, expressed a pattern of a short plateau and a decrease with reaching the state of memory overload, indicating that pupil size and theta possibly have similar neural mechanisms. Based on the described above triphasic pattern of pupil size temporal dynamics, we concluded that cognitive overload causes physiological systems to reset, and to release effort. Although memory capacity limits were exceeded and effort was released (as indicated by pupil dilation), alpha continued to decrease with increasing memory load. These results suggest that associating alpha with the focus of attention and distractor suppression is not warranted.

Revealing visual working memory operations with pupillometry: Encoding, maintenance, and prioritization

“…Whenever task demands become too high, observers experience "overload" and give up on the task. Giving up is accompanied by drops in pupillary dilation indicating reduced exerted effort (Granholm et al, 1996;Kahneman, 1973), which is in line with inverted-U patterns in frontal and parietal activity with increased task load (Kosachenko et al, 2023;Leung et al, 2004;Linden et al, 2003) (also see Ma et al, 2014). Such pupillary dilation effects during maintenance may be informative regarding the format of VWM representations (i.e., slots (Luck & Vogel, 1997, 2013Zhang & Luck, 2008) vs. resources accounts (Bays & Husain, 2008;Ma et al, 2014)).…”

“…Such pupillary dilation effects during maintenance may be informative regarding the format of VWM representations (i.e., slots (Luck & Vogel, 1997, 2013Zhang & Luck, 2008) vs. resources accounts (Bays & Husain, 2008;Ma et al, 2014)). Although direct studies into this issue remain scarce, some work has provided indirect indications into this debate without strong support for either account (see Kosachenko et al, 2023;Kursawe & Zimmer, 2015;Zhou et al, 2022). Targeted studies are necessary to reveal how pupillometry may inform about the format of VWM representations.…”