Midfrontal Theta Activity Is Sensitive to Approach-Avoidance Conflict

Abstract

Midfrontal theta (FMθ) in the human EEG is commonly viewed as a generic and homogeneous mechanism of cognitive control in general and conflict processing in particular. However, the role of FMθ in approach-avoidance conflicts and its cross-task relationship to simpler stimulus-response conflicts remain to be examined more closely. Therefore, we recorded EEG data while 59 healthy participants (49 female, 10 male) completed both an approach-avoidance task and a flanker task. Participants showed significant increases in FMθ power in response to conflicts in both tasks. To our knowledge, this is the first study to show a direct relationship between FMθ and approach-avoidance conflicts. Crucially, FMθ activity was task dependent and showed no cross-task correlation. To assess the possibility of multiple FMθ sources, we applied source separation [generalized eigendecomposition (GED)] to distinguish independent FMθ generators. The activity of the components showed a similar pattern and was again task specific. However, our results did not yield a clear differentiation between task-specific FMθ sources for each of the participants. Overall, our results show FMθ increases in approach-avoidance conflicts, as has been established only for more simple response conflict paradigms so far. The independence of task-specific FMθ increases suggests differential sensitivity of FMθ to different forms of behavioral conflict.SIGNIFICANCE STATEMENT FMθ is well established as an indicator for cognitive conflict in tasks involving simple stimulus-response conflicts. However, we do not yet know about its role in more complex forms of goal ambivalence, such as approach-avoidance conflicts. Thus, we implemented an approach-avoidance task and a flanker task to investigate FMθ in response to simple as well as more complex response conflicts. To our knowledge, this is the first study to show a direct relationship between FMθ and approach-avoidance conflicts. Although the transient FMθ increase is similar to that induced in a simple response conflict task, individual FMθ responsiveness to these two forms of conflict were independent of each other, suggesting intraindividual differences in the sensitivity of FMθ to different forms of behavioral conflict.

Keywords: FMT; approach avoidance; conflict processing; flanker task; midfrontal theta.

Figure 1.

Behavioral data. A, B, The mean approach/avoidance ratios and the mean decision times in the approach–avoidance task. C, The mean reaction times in the flanker task. Error bars indicate SEM.

Figure 2.

Conflict-related time-frequency activity at FCz. A, The differential activity for conflict minus no-conflict trials in the Flanker task as well as the topography for the early time window (146–434 ms) and the late time window (436–724 ms). B, The corresponding differential activity and the topography for the approach–avoidance task.

Figure 3.

Scatter plots for the relationships of conflict related activity within and across tasks and time windows. A, The scatter plots for the conflict-related time-frequency activity at FCz. B, The scatter plots for the conflict-related component activity.

Figure 4.

Exemplary results of the GED for two participants. The midfrontal template was used to determine frontomediality (R² > 0.4). The component topographies belong to the strongest component for each of the two exemplary participants. The eigenvalues of all components (eigenvectors) that were extracted are displayed. Colored dots indicate midfrontal components. The line of p = 0.05 displays the significance threshold that was computed with the permutation tests.

Figure 5.

Component time-frequency activity across all midfrontal components for the Flanker task and the approach–avoidance task.

Figure 6.

Conflict-related FMθ power at FCz per approach–avoidance condition for the late time window (436–724 ms). The data are grouped by individual conflict classification.