More evidence for a dual-process model of conditional reasoning

Area of Science:

• Cognitive Psychology
• Human Reasoning
• Decision Making

Background:

• Previous research suggested deductive inferences are governed by a single probabilistic model.
• The dual-process model of conditional reasoning proposes two distinct strategies: logical (counterexample-based) and probabilistic.

Purpose of the Study:

• To investigate whether deductive reasoning employs a single probabilistic model or multiple strategies.
• To examine the role of individual differences in abstract reasoning competence on strategy selection.

Main Methods:

• Two studies presented participants with affirmation-of-the-consequent inferences varying in statistical properties.
• Individual response patterns were analyzed to identify distinct reasoning strategies.
• A third study examined explicitly probabilistic inferences.

Main Results:

• Evidence supported the dual-process model, revealing two distinct reasoning strategies.
• Use of the counterexample strategy correlated with higher abstract reasoning competence.
• Explicit presentation of counterexamples facilitated the counterexample strategy.
• Higher abstract reasoning competence was linked to more accurate probabilistic inferences.

Conclusions:

• Deductive inferential reasoning is not explained by a single, unitary process.
• Individual differences in strategy use are crucial for understanding reasoning processes.
• Abstract reasoning ability plays a significant role in deductive inference accuracy.