Post 1: Shape Similarity Studies

I spent the initial part of my research analyzing data collected in the spring. Data was collected from 25 participants aged 18-22 years who signed up through SONA and were given credit towards an introductory psychology class as compensation. All subjects were tested for color blindness and found to have normal color vision. As mentioned in the abstract and background information, the study examined how shape similarity between a target and a salient distractor and the distance between the two affect visual selective attention using a target-decoy search task.

These experiments use either a target-decoy search task, or a context singleton search task. In the target-decoy task, 12 items are presented in a circular array with one item being the target item, one being the decoy item, and the remaining 10 being filler items. The items are arranged in positions corresponding to the hour positions on an analog clock face. The target was the letter ‘T’ orientated normally or upside down, and possessed a unique color. The decoy was either the letter ‘L’ or the letter ‘C’ and possessed a different unique color. The target and decoy colors were selected from two values: orange and green. The filler items were a randomly selected from target and decoy items and were gray. Subjects were asked to focus on a point at the center of the array and use their peripheral vision to identify the orientation of the target, using numpad key ‘1’ if the target was upright and ‘2’ if the target was upside-down.  MANIPULATION OF DISTANCE

Two sets of data were collected, analyzed separately and juxtaposed to determine the influence of shape similarity and distance on attentional selection. The first set of data was the measured event-related potentials (ERP) from the EEG data.  The ERP components of interest were the N2pc (as an index of competition), the Ptc (as an index of selectivity) and SPCN (an index of visual short-term memory processing). Upon gaining proficiency in using the software, BrainVision’s Analyzer 2 and IBM’s SPSS, I put each subject’s data through a series of transformations.TRANSFORMATIONS. Each subject’s segmented data was compiled into a grand average and mean peak information was exported to SPSS. The second set of data consisted of the behavioral information collected which included the relevant response keys, reaction times, and codes referring to the specific target-decoy arrangement responded to. This data was imported into SPSS.

The results of this study seem to oppose what was previously found. A possible cause may be poorly calibrated luminance values for the stimulus items. We are awaiting equipment and software to better measure the luminosity and to determine the validity of our results.