Performance in the Iowa Gambling Task in healthy aging and mild cognitive impairment

Area of Science:

• Cognitive psychology
• Neuroscience
• Computational modeling

Background:

• The Iowa Gambling Task (IGT) assesses complex decision-making.
• Performance differences exist between younger adults, healthy older adults, and those with mild cognitive impairment (MCI).
• IGT performance is influenced by risk preference, sensitivity to rewards/punishments, and memory, complicating interpretation.

Purpose of the Study:

• To utilize a computational cognitive model to identify specific cognitive mechanisms underlying IGT performance differences.
• To compare decision-making strategies in healthy older adults versus individuals with MCI.
• To examine decision-making across a wide age range in healthy adults.

Main Methods:

• Fitting a computational cognitive model to IGT data.
• Experiment 1: Compared healthy older adults (N=90) with individuals with MCI.
• Experiment 2: Compared healthy adults across all ages (N=1,645).

Main Results:

• Healthy older adults exhibited higher learning rates (recency bias), perseverative strategies, and greater sensitivity to gains.
• Individuals with MCI demonstrated slower learning rates but no significant differences in task strategy compared to healthy controls.
• Distinct behavioral profiles were observed between healthy aging and MCI on the IGT.

Conclusions:

• Computational modeling successfully attributed behavioral differences on the IGT to specific cognitive mechanisms.
• Findings highlight distinct cognitive alterations in healthy aging and early stages of cognitive decline (MCI).
• Results offer insights into impaired decision-making in the initial phases of cognitive decline, relevant for understanding dementia progression.