Range effects of an irrelevant dimension on classification.

Area of Science:

• Cognitive Psychology
• Perceptual Psychology

Background:

• Orthogonal classification tasks, or filtering tasks, involve ignoring irrelevant stimulus attributes, which typically slows response times compared to univariate tasks.
• Previous research indicates that increasing the range of irrelevant variation amplifies this performance slowing.

Purpose of the Study:

• To investigate how the range of irrelevant trial-to-trial variation impacts response times in pitch/loudness classification tasks.
• To determine the underlying mechanisms of Garner interference, specifically whether it arises from filtering difficulties or stimulus processing.

Main Methods:

• Experiments utilized pitch and loudness stimuli with varying levels of relevant and irrelevant attributes.
• Response times were analyzed using sequential analyses to examine the influence of trial sequence and stimulus sets.
• The study compared performance on trials with repeated versus changed responses.

Main Results:

• Experiment 1 confirmed that larger irrelevant dimension ranges slow responses, with sequential analyses revealing dependencies on both sequence and stimulus set.
• Experiments 2 and 3 demonstrated that irrelevant pitch variation slowed loudness categorization only on trials with repeated responses.
• Performance was unaffected by irrelevant variation when the response changed.

Conclusions:

• The interaction between response repetition and irrelevant variation suggests Garner interference is not due to filtering deficits.
• Slowed performance in orthogonal classification tasks is attributed to the processing of successive stimulus differences.
• Sequential analyses are crucial for uncovering nuanced cognitive processes beyond simple data averages.