Brain Imaging Investigation of the Neural Correlates of Observing Virtual Social Interactions

The overall goal of the following experiment is to investigate the neural correlates of observing social interactions by using virtual characters interacting in social context in conjunction with FMRI recording, this is achieved by creating whole body animated characters that interact with each other as hosts and guests in a business-like setting and inanimate characters that serve as control conditions. Next brain imaging data are recorded while participants observe virtual social interactions and then data are analyzed in order to identify brain regions that are part of the so-called social cognition network. Typical brain regions sensitive to observing and assessing social interactions include the superior temporal sulcus, lateral and medial frontal cortex and the amygdala Because it allows investigation of the neuro correlates of social cognition.

This matter might contribute to understanding possible causes of deficits in social behavior, which are observed in such clinical condition as social phobia and autism. The main advantage of this technique over existing methods, such as those involving facial stimuli, is that by using whole body animated characters, it's possible to incorporate manipulations that more closely resemble real life social interactions such as formal physical touch and handshake. This is something that's not possible with just facial stimuli.

The task consists of a series of ten second animated videos of nonverbal guest host interactions. In a business setting, the subject views the guest being greeted by a host, which is the social interaction condition or by a cardboard cutout of a host, which is the no social interaction or control condition. The host may display behaviors that are inviting to further social interaction called the approach, condition, or behaviors that may indicate lack of interest in further interaction being the avoid condition.

In addition to the basic type of behaviors displayed by the host, investigation of the effects of formal physical touch on behavior can be included. For instance, in half of the trials, the characters may shake hands as part of the greeting protocol, whereas in others, they do nothing. This manipulation may have different significance depending on the cultural background of the viewer.

Other manipulations may involve changing the viewer's perspective from personal to impersonal. By displaying the words, me or other, this manipulation allows identification of responses that are modulated by personal engagement. The alternation between these two perspectives can be queued at the beginning of each trial.

Videos can be followed by various rating screens. For example, asking the subject to rate the host on competence, trustworthiness, and interest in doing business on a five point scale with zero being not at all, and five being very much. These ratings should be counterbalanced across trials to avoid order effects.

Additionally, the characters in videos should be counterbalanced for displayed behavior, ethnic background, shirt, color, and hairstyle. Basic aspects that may influence social interactions such as attractiveness should also be controlled to ensure that there are no systematic differences between trial categories. Each run should begin with six seconds of a fixation to allow stabilization of the MR signal condition or trial order should also be randomized within each block and counterbalance between participants to avoid stimulus, predictability and order effects.

An inter trial interval of eight seconds follows each movie trial and the end of each run. Also to avoid bias data loss, ideally the condition should be equally represented in each block according to each manipulation. When the subject arrives, be sure to obtain written and informed consent and to screen for MRI safety.

Also have the subject complete questionnaires that assess personality traits and current state of mind, which may influence participants'responses during the study prior to the scan. Inform the participant in detail of the scan procedures and give specific instructions for the behavioral task to avoid discomfort and to create increased familiarity with the task. Also, give the participant an abbreviated practice run.

Now bring the subject into the scanning room and instruct him or her to lie supine on the scanning bed. Provide air protection as well as isolation headphones for communication. During the scan, provide additional cushioning for the head to ensure comfort during the scan and minimize movement.

Also, the non-adhesive side of a length of tape may be wrapped lightly around the participant's forehead to further minimize head movement to assess automatic implicit psychophysiological responses to emotional and neutral pictures In addition to the explicit ratings, skin conductance responses can also be recorded during the FMRI scanning. Finally, position the subject's right hand comfortably on the response box and check to make sure the buttons are working properly. Place an emergency stop button nearby so that the subject may indicate any urgent need to stop the scanner before starting data collection.

Assure that the subject can see the screen projection for stimulus presentation.Clearly. First, acquire a high resolution T one weighted structural image such as an SPGR or MP rage. Next, set up your functional runs to allow full brain coverage.

In our studies, we acquire a series of 28 functional slices with a four by four by four millimeter voxel size acquired axially. Using an echo plana pulse sequence with a TR of 2000 milliseconds, a TE of 40 milliseconds and a 256 by 256 millimeter field of view, inform the subject that the functional imaging is about to begin and remind him or her of the tasks that will be run. Now, be sure the scanner starts synchronously with the stimulus presentation.

When scanning is complete, help the participant out of the scanner and be sure to thank him or her for participating in the experiment. Once collected, data can be analyzed using any FMRI processing software package, such as statistical parametric mapping, commonly known as SPM. Our lab uses this in combination with in-house mats.

Lab based tools pre-processing in our group involves typical steps, quality assurance, image alignment, motion correction, co-registration, normalization, and spatial smoothing with an eight millimeter kernel. When using SPM, the general linear model is implemented to assess the fit of the data recorded for each condition of interest to a predetermined hemodynamic response function. Alternatively, the data from each condition can be selectively averaged to view the raw FMR signal associated with each condition with no predetermined assumptions about the shape of the hemodynamic response function.

In this case, differences in response can be assessed on a time, point by time, point basis, and contrast of interest reveal changes compared to pre stimulus, baseline, and differences between one condition when compared to another condition. These analyses can be used separately or complimentarily. Data analysis can combine voxel wise and region of interest approaches to compare brain activity associated with the conditions of interest such as social interaction versus no social interaction.

Whole brain voxel wise analysis will produce statistical maps that identify larger networks of brain regions associated with the task and ROI Analysis will allow for targeted investigation of the response in brain specific regions identified a priori as being part of the social cognition network. Region of interest analysis can also be used to extract the FMRI signal for illustration purposes. Comparison of social interaction versus no interaction control trials revealed activity in typical social cognition brain regions.

Here we see a functional map displaying activations in the superior temporal sulcus as well as the lateral prefrontal cortex. The color bar indicates the gradient of TValue of the activation map. The line graphs illustrate the time courses of the FMRI signal extracted from each brain region's ROIs for each trial type and time point.

Here we see activation in the amygdala, again with the ROI time course, and finally, the response in the medial prefrontal cortex. All data are based on data from 15 participants reflecting brain activity time locked to the onsets of approach and avoid behaviors While doing this procedure, it's important to use ecologically valid stimuli, which is properly counterbalanced and timed. After watching this video, you should have a good understanding of how to use whole body animated virtual characters.

In conjunction with FMRI recordings to investigate the New York Correlates of observing social interactions in both healthy and clinical participants.