Aversive Events as Positive Reinforcers: An Investigation of Avoidance and Safety Signals in Humans

Ioannis Angelakis, Jennifer L. Austin

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Introduction It is well established that stimuli correlated with the absence of aversive events may come to signal periods of safety from those events. In previous research, stimuli that signaled periods of safety were largely neutral in nature. In the present investigation, we examined whether aversive events also could be conditioned as safety signals. Methods Seven university students were initially exposed to a 72 dB tone, which all rated as highly aversive. Next, they were required to play a computerized game in which mouse clicks on a map produced points (treasures) or point losses (bombs). Participants were instructed that bombs could be disabled by pressing a foot pedal, although pedal presses pro- duced the aversive tone. In the next phase of the game, no bombs were programmed. During some sessions, pedal presses produced the tone and in other sessions they did not. Results Rates of pedal pressing maintained when the sound was produced; however, pedal pressing eventually ceased in sessions in which it was absent. Follow-up measures indicated that 6 of the 7 participants rated the tone as aversive at the end of the study. Discussion These findings suggest that the aversive stimulus functioned as a positive reinforcer, given that participants chose to produce the sound even when it did not abate aversive stimulation.
    Original languageEnglish
    Pages (from-to)627-635
    Number of pages9
    JournalPsychological Record
    Volume65
    Issue number4
    DOIs
    Publication statusPublished - 1 Dec 2015

    Keywords

    • Aversive stimuli
    • Avoidance
    • Safety signals
    • Two-factor theory

    Fingerprint

    Dive into the research topics of 'Aversive Events as Positive Reinforcers: An Investigation of Avoidance and Safety Signals in Humans'. Together they form a unique fingerprint.

    Cite this