Behavioral and EEG Measures Show no Amplifying Effects of Shared Attention on Attention or Memory

Abstract

Shared attention experiments examine the potential differences in function or behavior when stimuli are experienced alone or in the presence of others, and when simultaneous attention of the participants to the same stimulus or set is involved. Previous work has found enhanced reactions to emotional stimuli in social situations, yet these changes might represent enhanced communicative or motivational purposes. This study examines whether viewing emotional stimuli in the presence of another person influences attention to or memory for the stimulus. Participants passively viewed emotionally-valenced stimuli while completing another task (counting flowers). Each participant performed this task both alone and in a shared attention condition (simultaneously with another person in the same room) while EEG signals were measured. Recognition of the emotional pictures was later measured. A significant shared attention behavioral effect was found in the attention task but not in the recognition task. Compared to event-related potential responses for neutral pictures, we found higher P3b response for task relevant stimuli (flowers), and higher Late Positive Potential (LPP) responses for emotional stimuli. However, no main effect was found for shared attention between presence conditions. To conclude, shared attention may therefore have a more limited effect on cognitive processes than previously suggested.

Figure 1

Performance on the flower counting task. (a) Accuracy rates on the flower counting task and; (b) P3b topography (µv) in the window of [350:600]ms after the presentation of target stimuli in the alone (left) and shared (right) viewing conditions; (c) Stimulus-locked average activity (in µv) in the time window of [−100:800]ms. ERP data was calculated separately for the two viewing conditions (alone, solid lines; shared, dashed lines) and for the different stimulus types (target - flowers, green; non-target - neutral IAPS, black); Shaded error regions represent standard deviation errors (using the ShadedErrorBar MATLAB function). ERPs were computed across Pz CPz POz; (d) P3b mean activity in the 350–600 ms window, for each condition: target (flowers, green) and non-target (neutral IAPS, black) stimuli, viewed alone (left) and under shared attention (right; dashed) conditions. Error bars represent standard error of mean (SEM).

Figure 2

Shared attention effects for non-target emotional stimuli. On the flower counting task: (a) Mean LPP activity (in µv) in the time window of [400–800]ms averaged across the 3 electrode locations: POz, Pz and CPz; (b) Mean ERP activity in the time window of [−100:1000]ms (stimulus-locked) for POz, Pz and CPz ^*; (c); LPP topography (µv) in the [400–800]ms window following the presentation of non-target stimuli in each condition, and; on the memory task: (d) Performance sensitivity (d’) for the recognition task. Error bars represent standard error of mean (SEM). *For shaded error regions see Supplementary Fig. 1.

Figure 3

Alpha/mu band power (8Hz-13hz) during the flower counting task. (a-b) Logarithmic scale for the averaged activity within the 8–13 Hz range in locations C3, C4 (mu activity, panel (a), and in locations O1 O2 and Oz (alpha activity, panel (b). Error bars represent standard error of mean (SEM). (c) Topographic maps of power amplitude (µv²) in the 8–13 Hz range.

Figure 4

Experimental setup during performance of the flower counting task, in the alone (top) and shared (bottom) conditions. In the shared condition, a dual EEG setup was used, and amplifiers were synched using a ‘daisy-chained’ connection, which sent the information from both amplifiers to the experimental computer.

Figure 5

Trial procedure in the flower counting task. The sequence of trials in each block used in this task.