The use of Artificial Intelligence (AI) in the development of psychological intervention is in
the early stages and further research is needed to validate the efficacy of its impact on
improving mental health. Tess is a customizable AI chatbot that offers scalability of
emotional support services. This study includes a randomized controlled trial that
examined the use of Tess, the chatbot, vs. a self service psychoeducation intervention
delivered in Argentina.
The main goal was to evaluate the effectiveness of the chatbot during an eight week
period to reduce symptoms of depression and anxiety. The sample was composed of
University students ages 18 to 33, of which 87.2% were female. 50 participants completed
the depression measure, Patient Health Questionnaire (PHQ-9, Kroenke, Spitzer, &
Williams, 2001) ,and 63 completed the anxiety measure, Generalized Anxiety Scale
(GAD-7 - Newman, et al., 2002).
A t-test analyzed for independent samples (experimental group and control group)
indicated that there were no statistically significant differences between the study groups
for depression (t(48) = 1.47; p = .089) and anxiety scores (t(61) = .87; p =.388). However, a t-test for related samples showed statistically significant differences for depression (t(26) =2.47; p = .020) and anxiety scores (t(26) = 2.15; p = .041) from pre- to post- intervention in the experimental group, and no significant differences were observed for the control group for anxiety(t(22) = 1,00; p = .327) and depression (t(29) = . 413; p = .683). Preliminary results indicate that using Tess is a promising tool for reducing symptoms of depression and anxiety in University students in Argentina.