Department of Psychology, University of Regensburg, Regensburg, Germany.
Department of Information Science, Nara Institute of Science and Technology, Ikoma, Japan.
JMIR Form Res. 2024 Oct 15;8:e55234. doi: 10.2196/55234.
Cognitive behavioral therapy (CBT) is a valuable treatment for mood disorders and anxiety. CBT methods, such as cognitive restructuring, are employed to change automatic negative thoughts to more realistic ones.
This study extends on previous research conducted by the authors, focused on the process of correcting automatic negative thoughts to realistic ones and reducing distress and anxiety via CBT with a virtual agent. It was aimed to investigate whether the previously applied virtual agent would achieve changes in automatic negative thoughts when modifications to the previous experimental paradigm are applied and when user motivation is taken into consideration. Furthermore, the potential effects of existing participant knowledge concerning CBT or automatic thoughts were explored.
A single-group, 2-session experiment was conducted using a within-group design. The study recruited 35 participants from May 15, 2023, to June 2, 2023, via Inter Group Corporation, with data collection following from June 5 to June 20, 2023, at Nara Institute of Science and Technology, Japan. There were 19 male and 16 female participants (age range: 18-50 years; mean 33.66, SD 10.77 years). Participants answered multiple questionnaires covering depressive symptomatology and other cognitive variables before and after a CBT session. CBT was carried out using a virtual agent, who participants conversed with using a CBT dialogue scenario on the topic of automatic negative thoughts. Session 2 of the experiment took place 1 week after session 1. Changes in distress and state anxiety were analyzed using a Wilcoxon signed-rank test and t-test for paired samples. The relationships of motivation with cognitive changes and distress or anxiety changes were investigated via correlation analysis. Multiple linear regression was used to analyze the potential predictive qualities of previous knowledge of CBT and automatic negative thoughts regarding outcome measures.
Significant reductions in distress (all P<.001) and state anxiety (all P<.003) emerged throughout the first and second experimental sessions. The CBT intervention increased participants' recognition of their negative thinking and their intention to change it, namely their motivation to change it. However, no clear correlations of motivation with changes in distress or anxiety were found (all P>.04). Participants reported moderate subjective changes in their cognition, which were in part positively correlated with their motivation (all P<.007). Lastly, existing knowledge of CBT did not predict reductions in distress during the first session of the experiment (P=.02).
CBT using a virtual agent and a CBT dialogue scenario was successful in reducing distress and anxiety when talking about automatic negative thoughts. The promotion of client motivation needs to be critically considered when designing interventions using CBT with a virtual agent, and further experimental investigations on the causal influences between motivation and outcome measures need to be conducted.
认知行为疗法（CBT）是治疗情绪障碍和焦虑症的有效方法。CBT 方法，如认知重构，用于将自动产生的消极想法转变为更现实的想法。
本研究扩展了作者之前的研究，重点研究了通过虚拟代理进行 CBT 以纠正自动消极想法、减轻痛苦和焦虑的过程。目的是探讨当应用先前的实验范式修改并考虑用户动机时，先前应用的虚拟代理是否会导致自动消极想法的变化。此外，还探讨了现有参与者对 CBT 或自动思维的知识的潜在影响。
采用单组、2 次实验的组内设计进行。该研究于 2023 年 5 月 15 日至 6 月 2 日通过 Inter Group Corporation 招募了 35 名参与者，数据收集于 2023 年 6 月 5 日至 6 月 20 日在日本奈良先端科学技术大学进行。参与者中有 19 名男性和 16 名女性（年龄范围：18-50 岁；平均 33.66，SD 10.77 岁）。参与者在 CBT 会话之前和之后回答了多个涵盖抑郁症状和其他认知变量的问卷。使用虚拟代理进行 CBT，参与者通过关于自动消极想法的 CBT 对话场景与虚拟代理进行对话。实验的第 2 次会议在第 1 次会议后 1 周进行。使用 Wilcoxon 符号秩检验和配对样本 t 检验分析痛苦和状态焦虑的变化。通过相关分析研究了动机与认知变化和痛苦或焦虑变化的关系。使用多元线性回归分析了先前对 CBT 和自动消极思维的知识对结果测量的潜在预测质量。
在第 1 次和第 2 次实验会议期间，痛苦（均 P<.001）和状态焦虑（均 P<.003）均显著降低。CBT 干预增加了参与者对消极思维的认识以及改变它的意图，即改变它的动机。然而，动机与痛苦或焦虑变化之间没有明显的相关性（均 P>.04）。参与者报告了对认知的适度主观变化，部分与他们的动机呈正相关（均 P<.007）。最后，CBT 的现有知识并不能预测实验第 1 次会议期间痛苦的减轻（P=.02）。
当谈论自动消极想法时，使用虚拟代理和 CBT 对话场景的 CBT 成功地减轻了痛苦和焦虑。在使用虚拟代理进行 CBT 时，需要批判性地考虑促进客户动机，并且需要进行进一步的实验研究以确定动机和结果测量之间的因果关系。
