ブックタイトル明星大学 心理学年報 第32号

概要

明星大学 心理学年報 第32号

Hayashi：Development and Testing of a New Indirect Attitude Measure for Pictorial Objects3Filtering Unconscious Matching ofImplicit Emotions (FUMIE) TestThe present study proposes two methods withwhich to measure attitudes toward pictures, andcompares thereliabilityand validityofthemeasurein Experiment 1. The first method is to apply theFiltering Unconscious Matching of Implicit Emotions(FUMIE) test to pictorial objects (Mori,Uchida, & Imada, 2008). Although the originalFUMIE test is a paper-and-pencil task implementedin collective form to measure attitudes toward asingle target word,thepresent studyused a versionof the original FUMIE test modified to measureattitudes toward a target picture.In a manner identicalto Mori et al.(2008),theFUMIE test in Experiment1 contains two types of experimental block,positiveandnegative,tomeasureattitudestowardatarget picture. In each block, a target picture, twoemotionally positive filler pictures, and two emotionallynegative filler pictures appear individuallyseveral times in random order. The participants’task is to judge the emotional valence of the picturesas theyappear,then indicate their choices bypressing the L key for positive and the S key fornegative pictures as quickly and precisely as possible;responselatencies for the pictures arerecordedin ms. In addition, in the instructions for eachblock, participants are asked to press the L key(positive key) for the target picture on positiveblocks and the S key(negative key) for the targetpicture on negative blocks, regardless of the originalemotional valenceofthetarget picture.That is,participants have to convert the emotional valenceof the target picture according to block type. If aparticipant has a positive attitude toward a targetpicture,the response latencyfor the target on positiveblocks would be shorter than the responselatency for the target on negative blocks. Sinceinconsistencyin a participant’s original attitudeandconverted response toward the target on negativeblocks is expected to make participants behavecautiouslythroughout theblock,participants wouldrespond slowlytoward filler pictures as well as thetarget.Therefore, the difference in averageresponse latencies for positive and negative blocksis expected to reflect the participant’s attitudetoward the target.However,theFUMIE test maynot beeffectiveatmeasuring attitudes toward a picture. As noted,participants in the present experiment are testedindividually in front of a monitor;it is possible toexpect that such conditions prompt participants tobe more focused on thetask compared with participantsin the original FUMIE test, carried out in acollective pencil-and-paper manner. Such intensifiedconcentration makes the conversion of thevalence of the target easier; thus, the differencebetween the average response latencies for positiveand negative blocks is expected to be smaller.The Serial Evaluative ConversionTask (SECT) and its Underlying ProcessesTherefore, the present study proposes a secondmethod, the Serial Evaluative Conversion Task(SECT), to measure attitudes toward a target picture.As shown in Figure 1, in a task trial, twodifferent emotional pictures appear on thescreen insuccession. The Participant’s task is to judge theconsistency of the emotional valence of these twopictures and press thecorrect key(corresponding to“consistent”or“inconsistent”) as quickly and preciselyaspossible.Thefirstpictureinatrial remainson the monitor for 1 s, and the second pictureremains on the screen until participants press thecorrect key.Thekeypresses and responselatencies(in ms) for the second picture are recorded. Trialsincluded in a block arecategorized as filler(50% ofall trials)or target trials(50% ofall trials).In fillertrials,extremelypositive or negative filler picturesappear in succession.In target trials,a filler pictureappears as the first picture and a target pictureappears as the second picture (Figure 1). As in theFUMIE test, this task has positive and negativeblocks to measureattitudes toward a target picture.In positive blocks,participants are to respond as ifthe target picture is positive, irrespective of itsoriginal valence;innegativeblocks,participantsareto respond as if the target picture is negative,irrespectiveof its original valence(Figure 1).As in the