Benefits of response time-extended multinomial processing tree models: A reply to Starns (2018)

Area of Science:

• Cognitive Psychology
• Psychometrics
• Mathematical Psychology

Background:

• The response time-extended two-high-threshold (2HT-RT) model is a specific instance of response time-extended multinomial processing tree models (MPT-RTs).
• A critique questioned the 2HT-RT model's ability to account for speed-accuracy trade-offs in recognition tasks.
• A discrete-state model with a race mechanism was proposed as an alternative.

Purpose of the Study:

• To clarify the utility of the 2HT-RT model as an example of MPT-RTs.
• To highlight the advantages of MPT-RTs as a flexible modeling framework.
• To demonstrate that MPT-RTs can accommodate and measure speed-accuracy trade-offs.

Main Methods:

• Utilizing response time-extended multinomial processing tree models (MPT-RTs).
• Binning response times into discrete categories for joint analysis.
• Applying MPT-RTs to psychological paradigms, including those with ceiling performance or ambiguous accuracy criteria.

Main Results:

• The 2HT-RT model serves as a valuable example within the broader MPT-RT framework.
• MPT-RTs offer a flexible and user-friendly approach for modeling.
• Certain MPT-RT designs can effectively capture and quantify speed-accuracy trade-offs.

Conclusions:

• MPT-RTs provide a robust method for jointly analyzing discrete choices and response times.
• The MPT-RT framework is applicable across various psychological tasks, even those with performance limitations.
• The proposed MPT-RT approach successfully addresses the challenge of modeling speed-accuracy trade-offs.