FaceSay™ Study Results
There have been three randomized controlled studies of FaceSay that have been published, either in a peer reviewed journal [1,2], or as a doctoral dissertation paper [3], as well as a published repeated baseline Case Study [4]. All of the studies have been done independently.
In addition, there are two unpublished studies. One is a multi-baseline study with 7 lower-functioning students, ages 7-11 done for an undergrad thesis. The students who improved the most on emotion recognition are the ones who completed the most FaceSay sessions. Another is an RCT (N=26) with 3-5 yo students, both TD (N=7) and HFA (N=19), done for a Master's thesis. The HFA students improved significantly relative to controls on Face Recognition, improving to the level of the TD participants.
This extensive empirical evidence is not yet FDA-grade evidence, but it is a good start and unmatched in several areas:
- The playground interaction results are an important breakthrough that has not yet been matched by any other tech intervention.
- In lab based measures, no other tech product or academic project has shown improvements in Emotion Recognition, Face Recognition and Theory of Mind, or even in two of these three.
- No other tech intervention has been shown to help such a wide range of students on the spectrum. In particular both students with above average IQ (HFA/Aspergers) and below average IQ (classic autism) have shown statistically significant improvements. However, the benefit is smaller for the students with below average IQ.
- Who will benefit the most? Our data so far indicate that HFA/Aspergers students ages 5-11 benefit the most. In addition, TD and HFA students 3-5 may benefit from Face Recognition improvements. While using reinforcement to facilitate enjoyment of the game, FaceSay is designed to allow the kids to make their own inferences (FaceSay never mentions emotion labels, for example); which puts FaceSay closer to a Montessori rather than a Behavioral approach. As might be expected, the Hopkins paper reports a positive correlation between IQ and benefit.
- What are the active ingredients? Empirically, we have some evidence [3] that the Follow the Face Game - the only one with facial expression training - is necessary for emotion recognition benefits.
- How frequently should the kids play? It's important to remember that the benefits have been seen in a relatively short time frame - 6-12 weeks, playing once or twice a week for 15-30 minutes, with the best results for twice a week.
- With FaceSay's inherent fidelity and appeal for the kids, an RCT may be more manageable than you think. The manual is operationalized for you in machine code. For average to high IQ students as young as 5, they won't need any training and will gladly play the games on their own (an observer is common protocol). Students with a lower-IQ may need some reinforcement while playing. You can use any preferred approach. For the UAB study, the students received hand-over-hand work, and cheerios reinforcers. For the Mutli-Baseline study, the researcher provided verbal reinforcement/encouragement.
- Any non-social computer game can be used as the control. In the Hopkins study, a drawing program, Tuxe Paint was used. In the Rice study, SuccessMaker was the control.
- You probably have a hypothesis or two in mind, but I'm happy to provide additional ideas. For example, it would be interesting to replicate Vivanti and Rogers' (2008) dyadic imitation task. FaceSay participants would likely show increased dwell times relative to controls for face vs non-face and eyes vs mouth; and this would correlate with improved performance on the imitation task.
- I will gladly provide free FaceSay licenses for the duration of the study.
- For international researchers, a small grant could fund a translation of FaceSay. Here are some quick tests I've created in Italian, German, Turkish, French, Spanish, Swedish, Portugese, Polish and Dutch.
[1] Hopkins IM, Gower MW, Perez TA, Smith DS, Amthor FR, Wimsatt FC, Biasini FJ., (2011). Avatar assistant: improving social skills in students with an ASD through a computer-based intervention. Journal of Autism and Developmental Disorders, 41(11):1543-55
[2] Rice, Linda Marie. Generalization of social skills following a computer based intervention for elementary school aged children diagnosed with autistic spectrum disorders [In preparation]. Journal of Autism and Developmental Disorders, 2015
[3] Perez Crawford, Trista, Improving Face and Emotion Recognition Skills of Children in Head Start with the use of a computer-based intervention. [in preparation]. Dissertation available at ProQuest, UMI Dissertations Publishing, 2012, Dissertation # 3525981
[4] Dalrymple KA, Corrow S, Yonas A, Duchaine B., Developmental prosopagnosia in childhood, Cognitive Neuropsychology, 2012;29(5-6):393-418
[2] Rice, Linda Marie. Generalization of social skills following a computer based intervention for elementary school aged children diagnosed with autistic spectrum disorders [In preparation]. Journal of Autism and Developmental Disorders, 2015
[3] Perez Crawford, Trista, Improving Face and Emotion Recognition Skills of Children in Head Start with the use of a computer-based intervention. [in preparation]. Dissertation available at ProQuest, UMI Dissertations Publishing, 2012, Dissertation # 3525981
[4] Dalrymple KA, Corrow S, Yonas A, Duchaine B., Developmental prosopagnosia in childhood, Cognitive Neuropsychology, 2012;29(5-6):393-418