Spontaneous eye blink rate (sEBR) has been found to be a non-invasive indirect measure of striatal dopamine activity. Dopamine (DA) neurons project to the prefrontal cortex (PFC) through the mesocortical dopamine pathway and their activity is implicated in a range of cognitive functions, including attention and working memory (WM). The goal of the present study was to understand how fluctuations in sEBR during different phases of a working memory task relate to task performance. Across two experiments, with recordings of sEBR inside and outside of a magnetic resonance imaging bore, we observed sEBR to be positively correlated with WM performance during the WM delay period. Additionally we investigated the non-linear relationship between sEBR and WM performance, and modeled a proposed Inverted-U-shape relationship between DA and WM performance. We also investigated blink duration, which is proposed to be related to sustained attention, and found blink duration to be significantly shorter during the encoding and probe periods of the task. Taken together, these results provide support towards sEBR as an important correlate of working memory task performance. The relationship of sEBR to DA activity and the influence of DA on the PFC during WM maintenance is discussed.

An open question in the working memory (WM) field is how information is kept online during the WM delay period. Maintenance of simple stimuli in WM is supported by connectivity between frontal and parietal brain regions. How does delay period activity and connectivity support WM of complex stimuli? Twenty-two participants completed a modified Sternberg WM task with complex stimuli and were told to remember either 2 (low-load) or 5 (high-load) scenes while 32- channel scalp EEG was recorded. During the 6-sec delay period 6 phase-scrambled scenes were presented, which served as interference. While increasing the WM load, particularly with complex stimuli, places a greater demand on attentional resources, interfering stimuli may hijack the available resources. This was confirmed in the examination of theta and alpha amplitude, as amplitude was reduced for the high WM load as compared with the low WM load across frontal, central, and parietal regions. Delay period connectivity was assessed with phase-locking value (PLV). We identified 3 supporting networks that facilitated performance for the low-load condition: 1) increased PLV between left frontal and right posterior temporal in the theta and alpha bands; 2) increased PLV between right anterior temporal and left central in the alpha and lower beta bands; and 3) increased PLV between left anterior temporal and left posterior temporal in theta, alpha, and lower beta bands for the low-load condition. These results suggest that these brain networks facilitated the low-load WM by filtering of interference and the use of verbal rehearsal during the delay period.Impact StatementAlthough, studies of working memory maintenance with simple stimuli have suggested a role of frontal-parietal networks in supporting maintenance, the current study suggests that maintenance of complex visual stimuli with interference present is supported by interhemispheric frontal-posterior temporal and intrahemispheric left temporal region connectivity. These networks support maintenance by filtering of the interfering stimuli, which facilitates the use of verbal rehearsal strategies during the delay period.