dc.description.abstract |
In this study, the correlations between EEG sources in multiple brain areas and the performance feedback in a motivation game were studied. EEG data were collected from 14 (6 male, 8 female) participants with 19 channels according to the standard 1020 system. The task consisted of solving as many questions as possible in a given time-limited (2 min.) round. The questions were simple arithmetic, symbolic, and verbal items from typical IQ tests and displayed on the computer screen. Each participant competed with 10 human opponents for two rounds each, based on the responses collected from the opponents before the EEG experiments. After the first round against each opponent, the participant was given a text feedback ("You win!", "You lose!", or "It’s a tie!"), depending on his/her score when compared to the opponent’s score. Next, the participant was allowed to compete with the same opponent in a second round. After band-pass filtering, the artifactual components were identified by ICA and eliminated. Based on reconstructed EEG signals, dipole sources were localized by using EEGLAB and the standard boundary-element model. Multiple regression analyses were performed with total normalized moment magnitudes in different brain areas as the independent variables. Dependent variables were performance changes such as the net score change in the second round with respect to the first round, the change in the attempted questions, and the change in only correct answers. The results show that when positive feedback ("You win!") was given, the associative visual cortex, orbitofrontal area and ventral anterior cingulate cortex became significantly correlated with net score changes and the changes in the correct answers. On the other hand, when negative feedback ("You lose!") was given, dorsolateral prefrontal cortex and dorsal anterior cingulate cortex were significantly correlated with the same performance changes. Additional areas were also found in different rounds. |
|