Sunk Cost Effect in Individuals with Autism Spectrum Disorder.

Junya et al. (2019) asked adults with and without autism to finish a money game.

The game let players quit or keep spending after they had already paid.

The team wanted to see if autistic players felt the "sunk-cost" pull to stay because of past money.

Autistic adults walked away from the game more often.

They showed a smaller sunk-cost bias than the control group.

In plain words, prior spending bothered them less when choosing to quit.

Brosnan et al. (2016) found autistic learners often think wrong math answers are right.

Both studies show weaker use of context: past money or past accuracy.

Rogé et al. (2011) saw the same pattern in moral tasks.

Their autistic sample ignored intent when blaming or forgiving others.

Together the papers paint one picture: people with ASD give less weight to historical cues.

Reed (2023) looks like a contradiction at first.

That study says prior verbal feedback hurts autistic kids when rules change.

The key difference is task type.

Phil tested holding onto old words; Junya tested holding onto old money.

Both show difficulty updating, but the sunk-cost task rewards letting go, so autistic performance looks "better," not worse.

When you teach vocational or money skills, expect clients to drop a task once cost tops gain.

Use clear, forward-looking prompts instead of "but you already paid" logic.

Pair this with self-monitoring checks from Brosnan et al. (2016) so clients verify their new choice is correct.