Detailed Notes on Fisher et al. Study and Preference Assessment Methods

  • Grid Setup for Pairings

    • Assign numbers to stimuli (1-10) on a datasheet.
    • Write down each stimulus with corresponding labels (e.g., stimulus one = candy bars).
    • Circling paired stimuli (e.g., stimulus five with stimulus two) indicates how to record selections.
    • Box should contain the number of the chosen stimulus for each pairing.

  • Single vs. Paired Stimulus Procedures

    • The study by Fisher et al. compares single stimulus (PACE method) to paired stimulus (termed forced choice).
    • The field favors the term "paired stimulus" for clarity.
    • Significant issue with single stimulus is undifferentiated responding leading to false positives.
    • False Positive Definition: Indicates that a stimulus is preferred, but reinforcement assessments do not support this.
    • Aim: Whether paired stimulus produces a more accurate hierarchy of preferences than undifferentiated single stimulus.

  • Study Participants and Methodology

    • Four individuals with developmental disabilities participated.
    • 16 stimuli used, presented singly (10 times each) for single stimulus.
    • For paired stimulus: present stimuli in pairs; each stimulus paired with every other stimulus.
    • Successful selection results in access for 5 seconds, while no response ends that trial.

  • Data Representation

    • Graphs compare outcomes from both methodologies.
    • The original article used different terms but were adapted to reflect current terminology.
    • Problems with Single Stimulus: Prone to undifferentiated responding, leading to false positive identifications.
    • Emerging Preference Hierarchy: Fewer stimuli considered highly preferred when using paired stimulus versus single stimulus.

  • Reinforcer Analysis

    • Follow-up study assessed which identified stimuli functioned as effective reinforcers.
    • Reinforcer assessments confirmed preferences identified from both methods, but those only identified through single stimulus did not function as reinforcers.
    • Shows paired stimulus method's higher sensitivity in identifying true preferences and reinforcer efficacy.

  • Calculating Percent Preference

    • Percent preference = (Number of approaches to an item / Total trials for that item) × 100%
    • Ensure equal trials across stimuli for accurate comparisons.
    • Graph values with stimuli on the X-axis and selection percentages on the Y-axis.
    • Example calculation:
    • If stimulus 1 (e.g., chips) was selected 2 out of 9 trials: (2/9) × 100% = approx. 22%.

  • Graphing Results and Hierarchical Display

    • A completed datasheet shows insights about preferences and how they were calculated.
    • Individual stimuli preferences visualized via a bar chart.
    • Higher bars indicate more preferred items that are also potential reinforcers.

  • Limitations of Paired Stimulus Methodology

    • Prone to false negatives (missing true preferences).
    • Can be time-consuming based on the number of stimuli.
    • Example: With 10 items, would need 45 trials, which is extensive.
    • The assessment is a demand, leading to potential problem behaviors post-selection.
    • Possibility of positional biases, where choices depend on item placement rather than preference.
    • Need to rotate item placements in trials to minimize impacts.

  • Strategies to Address Positional Bias

    • Limit items to mitigate time spent on assessments.
    • Present alongside or hold items to encourage preference-based selection.
    • Arrange items in configurations that reduce proximity influence.
    • If bias persists, consider switching to single stimulus or other duration-based assessment methods.

  • Barre & Colleagues Study

    • Investigated interventions for positional bias by presenting known preferred and non-preferred items.
    • Three methods explored:
    • Quality Training: Response correction through pairing with known non-preferred stimuli.
    • Magnitude Modification: Larger amounts of preferred items used to encourage selections of higher magnitude items.
    • Error Correction: Blocking the selection of non-preferred stimuli to reinforce correct preferences.
    • Goal: To ensure selections are based on preference rather than positional influences to refine preference assessment accuracy.