The unequal variance signal-detection model of recognition memory: Investigating the encoding variability hypothesis

Area of Science:

• Cognitive Psychology
• Psychophysics
• Memory Research

Background:

• The unequal variance signal-detection (UVSD) model is a key framework for understanding recognition memory.
• A psychological basis for the UVSD model's unequal variance assumption, specifically the encoding variability hypothesis, remains unverified.
• This hypothesis posits that greater variance in memory strength for old items stems from variable encoding increments.

Purpose of the Study:

• To empirically test the encoding variability hypothesis by manipulating conditions expected to increase encoding variability.
• To determine if increased encoding variability leads to greater memory strength variance for old items (σ_o) within the UVSD model.
• To investigate the relationship between encoding variability manipulations and estimates of memory strength variance and discriminability (d').

Main Methods:

• Three experiments were conducted manipulating encoding variability.
• Experiment 1: Varied item presentation duration (fixed vs. variable).
• Experiment 2: Manipulated attentional load via auditory one-back task with fixed vs. variable distractor intervals.
• Experiment 3: Used stimuli with high vs. low word frequency variance.
• Recognition memory performance and signal-detection parameters (σ_o, d') were estimated.

Main Results:

• Attempts to manipulate encoding variability did not significantly alter estimates of σ_o across the three experiments.
• The manipulations had a weak, inconsistent effect on subsequent recognition performance.
• Estimates of σ_o were positively correlated with estimates of d', suggesting σ_o may scale with overall memory discriminability.
• Successful manipulation of encoding variability proved difficult.

Conclusions:

• The findings challenge the direct link proposed by the encoding variability hypothesis between manipulated encoding variability and memory strength variance.
• Current experimental paradigms may be insufficient for effectively manipulating encoding variability in recognition memory tasks.
• Further research is needed to refine methods for testing the encoding variability hypothesis and its implications for the UVSD model.