PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • Eighteen healthy volunteers (6 males; mean age=21.1 years, Standard Deviation, SD=1.1 years) participated in the current study. All of the subjects were right-handed with normal or corrected-to-normal vision. Three of them were excluded from data analysis because of head movement during scanning (one participant) or because they were categorized as an outlier because the number of error trials was more than two SDs from the mean (one participants)[30]. An additional participant was categorized as an outlier because self-reports of pride were more than two SDs below the mean. Thus fifteen subjects were included in the final analysis. This study has been approved by the IRB at Southwest University. We had obtained appropriate ethics committee approval for the research reported, and all subjects gave written informed consent in our experiment. This experiment consisted of two sessions, a pre-scan behavioral test session and an fMRI scan session the following day. For session one, participants were required to engage in a standard time estimation task in which the time window changed according to participants’ accuracy; the task is designed such that the feedback depends on performance and yet results in 50% correct feedback [31]. Participants were told that the aim of this session was to test their time estimation ability and that they would be assigned to a two-person team with another stranger who had the same level of time estimation ability. The following day, they finished a team cooperation task with their partners (who were actually confederates of the investigator) in the fMRI experiment. Note that the real participants were always assigned to play the game in the fMRI scanner. For the fMRI session, at the beginning of each trial of the time estimation task a black fixation cross appeared on a gray background (1500 ms) and then turned red (500 ms) (Figure 1a), serving as a warning stimulus. Afterwards, the word “Go!” appeared on the screen (2000 ms maximum). The participants were told to press a button when they believed that 1000 ms had elapsed since stimulus onset, after which the “Go!” sign disappeared, resulting in a blank screen of random duration; the total duration of the “Go!” stimulus and the blank screen was fixed at random, integral multiples of the TR (i.e., 2000 ms, 4000 ms, and 6000 ms). Then, a “team feedback” stimulus appeared (2000 ms), indicating whether the team won or lost, represented by a smiling face and crying face, respectively, followed by another blank screen with random duration (2000 ms, 4000 ms, 6000 ms). Finally, “individual feedback” appeared (2000 ms), whereby information about the participant’s own performance and his/her partner’s performance was displayed under the words “You” and “Other”, respectively. A cross in a circle indicated poor performance and a check mark in a circle indicated good performance. Participants were told that team success depended on the average performance of their own reaction time (RTself) and their partner’s RT (RTother) according to the following rule: RT for team (RTteam)=(|RTself-1s|+ |RTother-1s|)/2. Thus, if the average RT was within the time window from 0 to 100 ms then the team won, otherwise, they lost. In fact, unknown to participants the team feedback was pseudo-randomly generated such that it indicated success on half of the trials and failure on the other half. However, for trials with RTs longer than 1800 ms or shorter than 400 ms, team feedback was always negative and individual feedback always indicated guilt, so as to make the feedback seem believable for extreme errors. All stimuli appeared in the center of the screen and were approximately 5 degrees of angle. According to our hypothesis, participants would attribute responsibility differently in accordance with team and individual feedback. The two feedback conditions together describe six conditions: high guilt (HG: the team failed because of the participant’s own bad performance), medium guilt (MG: the team failed because both members performed badly), low guilt (LG: the team failed because of the partner’s bad performance), high pride (HP: the team succeeded because of the participant’s own good performance), medium pride (MP: the team succeeded because both players performed well), and low pride (LP: the team succeeded because of the partner’s good performance) (see Figure 1b). The experiment was divided into four runs. In total, there were 108 trials for each condition of team feedback (team won vs. lost) and 36 trials for each of the six combinations of individual and team feedback. After participants finished the experiment they were asked to finish a questionnaire consisting of three parts. In the first part, they had to complete a 9-point questionnaire to rate their degree of happiness when they faced two different situations of team feedback and six situations of individual feedback. Also, they were required to rate their feeling of responsibility (pride and guilt) in relation to different types of individual feedback. “1” indicated very low pride or guilt when the question was related to responsibility attribution and indicated very unhappy when the question related to happiness. Conversely, “9” indicated very high pride/guilt and very happy, respectively. In the second part of the questionnaire they were presented with the following forced-choice question: “Which is more important for you, the success of your team or your own performance?” The third part contains eight questions from the subscale “dutifulness” of big five questionnaire. Since no significant result was found with regard to the score of “dutifulness”, the data of third part will no longer be discussed in the rest of this paper. Blood oxygen level dependent (BOLD) signals were measured using a 3.0 T Siemens MAGNETOM Trio scanner (Allegra; Erlangen, Germany) with a 12-channel head coil. Functional MRI data were acquired in four separate runs using a T 2-sensitive gradient echo planar imaging sequence covering the whole-brain (32 slices, slice-thickness 3.0 mm, Repetition Time (TR)= 2000 ms; Echo Time (TE)=29 ms, field of view (FoV): 220 mm2, matrix size: 64 × 64. An anatomical 3D dataset consisting of 176 slices was acquired between the second run and the third run (MDEFT sequence (Deichmann et al., 2004); TR= 1900 ms; TE =2.52 ms; Flip Angle=9°, voxel dimensions=1×1×1 mm3; FoV=250 mm2). Unlike the standard 1 s time-estimation task [31], in the fMRI component of the present study all of the feedback stimuli were presented pseudo-randomly. Therefore, participant accuracy was not meaningful in this experiment. Nevertheless, participants could modify their behavior on a given trial based on feedback received on previous trials. For this reason, we calculated the absolute value of difference in RT between consecutive trials (△RT) as a measure of error correction. In addition, correlation analyses were conducted among responsibility scores, happiness scores and △RT values. All the r-values reported here reflect Pearson correlation values. Imaging data were analysed using BrainVoyager QX (Brain Innovation, Maastricht, the Netherlands). Functional data were preprocessed to correct for slice scan time differences (using sinc interpolation), 3D motion artifacts (Trileaner sinc interpolation), linear drifts, and low-frequency non-linear drifts (high pass filter less than 2 cycles/time course). Functional data were then co-registered with the anatomical volume and transferred into standard stereotaxic space using Talairach normalization [32] and spatially smoothed with a 4 mm full width at half maximum Gaussian kernel. The statistical analyses were carried out using a voxel-wise General Linear Model (GLM) at the single subject level, based on design matrices that included the estimated 3D motion parameters obtained during pre-processing as well as predictors for all relevant task conditions (two events of team feedback, six events of individual feedback and the “Go!” stimulus). The reported group analyses were conducted using a random effects model. Note that only the time periods associated with team feedback and individual feedback were of interest. The analyzed time window of each event was 2s. And the box-car regressor of feedback stimulus was calculated throughout the entire duration of feedback presentation. The time of box-car regressors of all conditions were the same. The two conditions of team feedback (team loss and team gain) were submitted to a T-test, and the three pride conditions and three guilt conditions were submitted to ANCOVA analysis in Brain Voyager separately. The statistical group maps representing significant results were corrected for multiple comparisons using the false-discovery rate (q <.05, FDR) with 10 continuous voxels [33]. As an exploratory analysis, regions of interest (ROIs) were created based on the significant clusters of activation identified in voxelwise analyses of group level for the main effects of pride and guilt attribution (see results). The ROI was defined for each subject based on the peak coordinates of the results from the group level. Then, ROI analyses were performed by extracting parameter estimates (betas) from the GLM model and averaging across all voxels in the cluster for each subject. These beta values were further subjected to correlation analyses between the BOLD and behavioral data. For the pride and guilt conditions separately, correlations were performed across the three sub-conditions (high, medium, low) between the BOLD signal and △RT. We also computed differences in behavioral data (separately for RT, happiness scores and responsibility scores) between pairs of individual feedback conditions (e.g., HP and MP), and correlated these values with the corresponding differences in the BOLD signal for each of the ROIs, separately.
rdf:type