Shape bias in word learning: Insights from ASD

Source: psycnet.apa.org

<x

TL;DR

The story at a glance

Saime Tek and Letitia Naigles examine the shape bias—a key mechanism in how typically developing children learn object names—through the lens of autism spectrum disorder (ASD). They draw lessons from studies showing reduced shape bias in children with ASD and discuss implications for therapy. This review appears in Translational Issues in Psychological Science in March 2017, amid growing interest in language delays in ASD.[[1]](https://psycnet.apa.org/record/2017-10606-009)

Key points

Details and context

The shape bias was first documented by Landau, Smith, and Jones (1988, 1992) in typically developing toddlers; it reflects how children prioritize shape for rigid objects like "cup" or "bottle," ignoring superficial features. This bias accelerates learning because shape often correlates with category boundaries in man-made objects.

In ASD, prior work like Potrzeba, Fein, and Naigles (2015) and Tek et al. (2008) used matching tasks where children chose referents for novel words (e.g., "wug") between same-shape/different-color or same-color/different-shape options. Typically developing kids picked shape matches ~70-80% of the time; ASD kids performed at chance or preferred other features.[[3]](https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2015.00446/full)

The review synthesizes these experiments, noting variability in ASD possibly due to visual processing differences or narrower attention. It argues for targeted therapies building on emerging shape bias to improve generalization in word learning.

Key quotes

None reliably sourced from full text.

Why it matters

Gender differences—no, this work highlights a core word-learning mechanism atypical in ASD, informing why vocabulary growth lags despite intact perception.

For clinicians and parents, it suggests focusing interventions on perceptual cues like shape to speed noun acquisition in young children with ASD.

Watch for intervention trials testing shape bias training; outcomes could vary by ASD severity and verbal level.[[4]](https://www.researchgate.net/publication/314484465_The_shape_bias_as_a_word-learning_principle_Lessons_from_and_for_autism_spectrum_disorder)