Reading Guide for Strack, Martin, & Stepper (1988)

Conferences of Aug 30/31

You can download this article from the Psy 121 e-reserve list. Please print the article, read it using this guide to help you, and bring your copy to your lab section on

Aug 30/31 and again on Sept 6/7.

Summary

The introduction to this paper provides a clear description of the facial feedback hypothesis and of several procedures that have been used to demonstrate the influence of facial feedback. It emphasizes different explanations (cognitive and physiological) that have been forwarded to explain the evidence obtained in previous studies. It then describes a new procedure that covertly influences facial muscle activity, i.e., that the authors believe influences facial muscle activity without the participants’ awareness. The two studies described in this paper test whether this procedure modifies self-reported experiences of participants viewing Gary Larson cartoons.

Our task in conference will be to review the facial feedback hypothesis, to understand how the investigators implemented an empirical test of this hypothesis and the predictions they made. We will focus primarily on Study 1, although we will also review why the investigators ran Study 2. We’ll outline the procedures of these studies and identify the independent and dependent variables. An important part of this exercise will be to compare and contrast the procedures used in these studies with the procedures implemented in the experiment in which you participated during the first class session. We’ll discuss the findings of Study 1 of Strack et al. and think through what approaches we should take to evaluating the data collected during the first class.

The guide below is intended to help you read the article by providing definitions of terms therein that you might not know. It focuses particularly on issues of experimental design, procedure and a first-pass assessment of the findings, drawing especially on materials in the assigned reading from Chapter 1 of Gleitman et al. (pp. 13-33). Next week, we will undertake a more detailed examination of the data analyses in Study 1, aided by the assigned reading in the Appendix of Gleitman et al.

Introduction

Overall, I think the introduction to this paper is clear. The one exception is in the last sentence of the next-to-last paragraph on the first page (p. 768): “No agreement exists, however, on how people’s facial expression influence their emotional reactions” (italics mine). As the subsequent sentence in the paper suggests, how refers to the mechanisms underlying the effect, not to whether facial expressions increase or decrease emotional reactions.

Some terms, in the order in which they appear in the introduction:

hedonic value (p. 768): the degree to which a stimulus or event is perceived as pleasant/unpleasant; sometimes referenced to whether the stimulus/event generates approach or escape

explicit manipulation (p. 770): in this context, “explicit” means that the facial expression has been named by the experimenter (e.g., the participants have been asked to smile). However, participants can become explicitly aware of the nature or intent of a manipulation, such as a set of instructions to contract this or that muscle, even if it is not verbally labeled.

obicularis orbis, zygomaticus, risorius muscles: these will be identified in class. The latter two are muscles in the face that are involved in the production of a smile. Another important facial muscle is the corrugator supercilii muscle, involved in the production of a frown. All of these muscles occur in pairs, one on each side of the face.

Questions to consider:

  1. What is the facial feedback hypothesis? What does it predict, both in general and about the outcomes of the experiments reported in Strack et al.?
  1. How has this hypothesis been tested in the past, and what are some of the concerns raised about these previous tests?

Method, study 1

Questions to consider:

  1. What is the “cover story” used by the investigators to mask the intent of their manipulation? Did it seem to be effective? Cover stories are often necessary in psychology experiments. Why? What do you think about this aspect of psychology research?
  1. How did the investigators manipulate facial expression? How many conditions were there in the experiment? What were the independent variables? Were they manipulated between-groups or within-subjects?
  1. How were the affective responses of the participants measured? Specifically, what was the primary dependent variable?
  1. Why were the participants’ ratings of task difficulty measured?
  1. How do the procedures implemented in this study compare to those in the study run on Tuesday in class? Do you think any of the changes in procedure will influence the findings?

Evaluating the findings, study 1

  1. What were the predicted findings of this study?
  1. Looking at Table 1, were the predicted findings obtained?
  1. Does it look like the differences obtained in the ratings of cartoon funniness could be due to differences in task difficulty?

Study 2

  1. What additional questions are addressed in Study 2?
  1. What is the additional independent variable in this study?
  1. What is the additional dependent variable in this study?
  1. If we have time, we’ll take a look at the findings in Table 2. Were the findings of Study 1 replicated in Study 2?

Study 1 Design

Different cover story, associated with the tasks to be performed

Pen actually used in experimental positions to make ratings

Practice task, two other tasks preceded target task

Third condition: nondominant hand.

Target task: rating of funniness of 4 Gary Larson cartoons, presented individually

Difficulty assessed before cartoon ratings (in assn with second and third tasks)

Tasks self-paced?

  1. The investigators evaluated the effect of pen position on funniness ratings by performing a linear trend analysis. In class, I’ll review with you what this analysis is. The analysis we’ll run is the mixed model analysis of variance (ANOVA) described in the third paragraph of p. 772, and we’ll go over this analysis in some detail in class.
  1. The investigators examined the effect of pen position with the participant’s ratings of task difficulty “controlled for.” This means that the degree to which the effects of pen position on funniness ratings were mirrored by the effects of task difficulty on funniness ratings was assessed. This exemplifies an attempt to eliminate the “third variable problem” described on pp. 25-26 of Gleitman et al. In short, the investigators found that pen position and task difficulty did not have the same effects on funniness ratings.

Outline of Experiment 1

Run during first class, August 28

Materials:

Recordings of 4 musical excerpts (30 sec each?)

Individual student packets. 5 sheets

Cover sheet explaining purpose, mailstop id

4 “data” sheets, reiterating instrux and providing rating scales

? computer program to generate visual stimuli

Pens/pencils

Cover story: pre-testing to determine best method for holding pen “at the ready” for responding during a bimanual music tempo tracking task. Four music excerpts. Primary task: tapping in time with primary rhythm, alternating between left and right forefingers. In this pre-test, the tapping won’t be measured, but important to do the task. At the same time, determine whether there is an “O” and/or a “T” presented among the “X’s” on the screen.

Three pen-holding instrux: between teeth, in lips, behind right ear

Same for all four excerpts (between-group manipulation)

Assessments:

Task difficulty: for tapping, visual, pen-holding, overall

Task success: for tapping, visual, pen-holding, overall

Pleasantness

Tenseness

Recognize music?

Like the piece?

Primary analyses

Group (3, between-groups) x music piece (4, within-subjects) ANOVAs on

  1. pleasantness
  2. tenseness
  3. like the piece?

Correlation of each of the above DVs with overall difficulty, success