AI: Revolutionising Functional Behaviour Assessment?

Functional Behaviour Assessment (FBA) is a crucial process in understanding why certain behaviours occur. It involves carefully observing and analysing a person's behaviour within their environment to identify the triggers and consequences that maintain the behaviour. Traditionally, this has been a time-consuming and labour-intensive process, often relying on manual observation and data collection. But what if there was a way to make FBA more efficient, accurate, and readily available.

Enter artificial intelligence (AI).

Researchers are exploring how AI can revolutionise FBA, and the results are promising. Imagine a world where wearable sensors and video cameras seamlessly collect data on an individual's movements and interactions throughout the day. This isn't science fiction; it's happening now. Researchers are using these tools to capture a wealth of information about behaviour and its environmental context.

AI algorithms can analyse this data automatically, detecting and quantifying behaviours that might be difficult or time-consuming for humans to track manually (Plotz et al., 2012). This automated analysis can provide valuable insights into behavioural patterns and triggers.

One particularly exciting application of AI in FBA is the detection and classification of specific behaviours. Cantin-Garside et al. (2020) demonstrated the power of AI in identifying self-injurious behaviours (SIB). Their research showed that AI could detect and classify different types of SIB, such as hair pulling, with remarkable accuracy – reaching levels as high as 97%.

Think about the implications for clinical practice. Embedding AI-powered data collection into clinical settings could free up therapists' time and attention. Instead of focusing on manual data collection, therapists can dedicate their expertise to responding to other critical clinical behaviours and providing individualised care. Furthermore, automated data collection can minimise errors associated with manual tracking, ensuring more accurate and reliable data.

Perhaps the most significant benefit is the potential for more immediate and accurate detection and assessment of ongoing behaviour. This real-time insight allows for more timely interventions, potentially preventing challenging behaviours from escalating and improving overall outcomes.

While the integration of AI into FBA is still in its early stages, the possibilities are vast. As research continues to advance, we can expect to see even more innovative applications of AI in behaviour analysis, leading to more effective and personalised interventions for individuals in need. The future of FBA may very well be intelligent.

Reference:

• Doan, T., Sullivan, B., Koerber, J., Hickok, K., & Soares, N. (2024). Perceptions of Machine Learning among Therapists Practicing Applied Behaviour Analysis: A National Survey. Behaviour analysis in practice, 17 (4), 1147-1159.
• Jennings, AM. & Cox, DJ. (2024). Starting the Conversation Around the Ethical Use of Artificial Intelligence in Applied Behaviour Analysis. Behaviour analysis in practice, 17 (1), 107-122.