Infant Effects on Experimenter Fidelity: New Data

• In Event: Poster Session 12
  In Poster Session: PS 12 Topic - Methods, History, Theory

Abstract

A considerable literature has emerged documenting the impact of children’s characteristics on the care they receive (cf. Bell, 1968). However, surprisingly little research has focused on how child characteristics may impact experimental procedures in otherwise standardized experimental settings. In one recent study, two experimenters in a gaze following procedure were found to look longer at target objects as a function of infants’ temperament, suggesting that experimenters’ procedural fidelity was influenced by child characteristics (Dixon et al., 2018). In this report, we present fidelity data from a different study, using different experimenters and different experimental tasks.

Fifty-three infants (22 girls) visited the lab at M = 18.56 months (SD = 0.95 months). Mothers completed the Early Childhood Behavior Questionnaire (ECBQ; Putnam et al., 2006). Temperament measures derived from the ECBQ included three overarching superdimensions (surgency, negative affectivity, and effortful control) and 18 subdimensions. Infants were presented four trials of an inappropriate-object-substitution-type procedure as well as four trials of a Baldwin-type novel-word-learning procedure. For the object substitution task, the objects were highly familiar to children; these included a cup, brush, teddy bear, and car. For the novel-word-learning task, objects were unknown to children (e.g., pastry blend, tire gauge, garlic press). For each trial, experimenters were instructed to present the props for a 30-second “familiarization period;” although exact exposure time could be influenced by a number of factors, including experimenter differences as well as infant cooperation and interest/boredom.

Five experimenters ran babies through the experimental procedure. For simplicity, analyses are based on data from the three experimenters who handled 83% of the babies. As shown in Table 1, these three experimenters did not systematically differ from one another in terms of how long they familiarized infants to the props across the 8 trials. But for all three experimenters, familiarization times were shorter in the novel-word-learning task [F(1, 39) = 280.92, p = .000]. Infants’ lack of familiarity with the novel objects in the novel-word-learning task may have caused experimenters to move along sooner than they were supposed to. In addition, average familiarization times within each task were associated with infant temperament (Table 2), with the associations varying by task. Familiarization times with the novel objects in the novel-word-learning task were significantly shorter when children were high in surgency. In contrast, familiarization times with the familiar objects in the object-substitution task were significantly longer when children were high in negative affectivity.

These results raise the possibility that experimenters in experimental settings, perhaps especially because they are blind to child temperament status, may inadvertently adjust their procedural implementation in response to child characteristics when implementing “standardized” experimental protocols. Although efforts can be taken to prevent such effects, it is important to recognize that as human beings, experimenters will ultimately have to respond to child characteristics, and thus, may modify their task implementation in systematic ways that differentially assist or inhibit different classes of infants. Future experimental researchers may wish to include temperament instruments as standard protocol to test for experimenter fidelity. 

Authors

• Wallace E Dixon, East Tennessee State University
  Presenting Author

• Lauren Driggers-Jones, East Tennessee State University
  Non-Presenting Author

• Chelsea Robertson, West Liberty University
  Non-Presenting Author