|
|
ENQUIRE PROJECT DETAILS BY GENERAL PUBLIC |
| Project Details |
| Funding Scheme : | General Research Fund | ||||||||||||||||||||||||||||
| Project Number : | 14601120 | ||||||||||||||||||||||||||||
| Project Title(English) : | Effects of Emotion Lexicon Training for Counsellors on Counselling Process and Outcome | ||||||||||||||||||||||||||||
| Project Title(Chinese) : | 輔導員情感詞彙訓練對輔導過程和結果的影響 | ||||||||||||||||||||||||||||
| Principal Investigator(English) : | Prof CHUI , Tsz Yeung Harold | ||||||||||||||||||||||||||||
| Principal Investigator(Chinese) : | |||||||||||||||||||||||||||||
| Department : | Dept of Educational Psychology | ||||||||||||||||||||||||||||
| Institution : | The Chinese University of Hong Kong | ||||||||||||||||||||||||||||
| E-mail Address : | haroldchui@cuhk.edu.hk | ||||||||||||||||||||||||||||
| Tel : | |||||||||||||||||||||||||||||
| Co - Investigator(s) : |
|
||||||||||||||||||||||||||||
| Panel : | Humanities, Social Sciences | ||||||||||||||||||||||||||||
| Subject Area : | Psychology and Linguistics | ||||||||||||||||||||||||||||
| Exercise Year : | 2020 / 21 | ||||||||||||||||||||||||||||
| Fund Approved : | 812,500 | ||||||||||||||||||||||||||||
| Project Status : | Completed | ||||||||||||||||||||||||||||
| Completion Date : | 30-6-2024 | ||||||||||||||||||||||||||||
| Project Objectives : |
|
||||||||||||||||||||||||||||
| Abstract as per original application (English/Chinese): |
The exploration and processing of emotion in counselling and psychotherapy involve emotion labelling and meaning making of emotional experiences. Putting feelings into words, in particular, has been shown to lower limbic system activation and the experienced level of emotional distress. Counsellors’ ability to facilitate clients to name emotions is therefore a crucial skill. However, some collectivistic cultures, including the Chinese culture, discourage open expression and discussion of feelings, especially negative ones. Counsellors and clients therefore need to navigate the articulation of emotions within the confines of traditional customs. Besides cultural factor, individual differences in how counsellors themselves experience emotions, such as emotion reactivity, emotion regulation, and emotion expression, may also influence their readiness to help clients to explore feelings.
This proposal examines the effectiveness of an emotion lexicon training intervention for counsellors in relation to counselling process and outcome. In Study 1, emotion words from existing counselling session transcripts will be identified. Forty counsellors and counselling trainees will be recruited to rate the relevance and usability of these words, along with other emotion words, in the counselling context. The most highly rated 240 words will be selected as training material for Study 2.
In Study 2, 60 counsellor-client dyads will be recruited. Using a within-subject control design, counsellors will receive emotion lexicon training before three consecutive counselling sessions (i.e., 80 words each time) and no lexicon training before another three counselling sessions. The order of intervention and control conditions will be counterbalanced among counsellors. Clients will report psychological symptoms before each session and rate session quality and counsellor empathy after each session. Counsellors will also rate session quality after each session. Session recordings will be transcribed verbatim, and counsellors’ and clients’ emotion word used will be identified using the Linguistic Inquiry Word Count software (Pennebaker, Booth, Boyd, & Francis, 2015). The effect of emotion lexicon training will be investigated in terms of counsellors’ and clients’ emotion word use and counselling process and outcome variables. How the effectiveness of training varies with counsellors’ individual difference in experiencing emotions will also be explored.
To increase the impact of this research, emotion words used in sessions that received high versus low ratings of session quality and counsellor empathy will be identified. These words, together with case examples, will be incorporated in an evidence-based online self-learning package where counsellors in Hong Kong can learn about the effective and ineffective use of emotion words in counselling. 在輔導和心理治療之中，探索和處理情緒會涉及情緒表達及其所表達的意義。有研究顯示，透過言語表達感受能夠減低大腦邊緣系統的活躍和情緒困擾的程度。由此可見，輔導員能促進受導者說出情緒是一個重要的輔導技巧。然而，在一些集體主義文化的處境，包括中國文化，不鼓勵公開表達和討論感受，特別是負面的感受。所以輔導員和受導者需要在傳統習慣的限制之中有效表達情緒。除文化因素外，輔導員們如何體驗情緒，例如輔導員的情緒反應、調節和表達等個人因素都有可能影響他們能否協助受導者探索感受。 本研究旨在探究輔導員接受的情緒詞彙訓練對輔導過程和成效之關係。研究一，首先，透過現存的輔導對話所抄錄的文本摘取情緒詞彙。然後，邀請四十位輔導員及實習輔導員以輔導相關程度和可用性為情緒詞彙評分，最高評分的二百四十個詞語會作為研究二的訓練材料。 研究二，由六十對輔導員和受導者組成，使用受試者內對照設計，輔導員會在三次連續的輔導前接受情緒詞彙訓練（每次八十個詞語），而另外三次連續的輔導前則不會接受情緒詞彙訓練。所有輔導員皆以接受訓練及不接受訓練的研究次序作對照。受導者會在每次輔導前回答有關心理症狀的問題，每次輔導後也會對輔導質素和輔導員的同理心評分，而輔導員完成輔導後會為輔導質素評分。每次輔導都會被錄影和逐字抄錄，輔導員和受導者所用的情緒詞彙會用語文探索與字詞計算軟件辨認出來 (Pennebaker, Booth, Boyd, & Francis, 2015) 。透過輔導員和受導者的情緒詞彙的使用與輔導的過程和成效來檢視情緒詞彙訓練的影響，本研究也會探究輔導員體驗情緒的個人因素如何影響情緒詞彙訓練的有效程度。 為了增加此研究的影響力，那些在輔導質素和輔導員同理心中得到高或低評分的情緒詞彙會被辨認出來，這些情緒詞彙和個案例子將會收錄在一個以實證為本的網上自學套件上，讓香港的輔導員可以從中學習到如何在輔導中善用有效和認知低效的情緒詞彙。 |
||||||||||||||||||||||||||||
| Realisation of objectives: | Objective 1: To study the effects of pre-session emotion lexicon training, 64 counselor-client dyads were recruited (exceeding the 60 dyads originally proposed). The 64 dyads were evenly divided between the experimental and control conditions to test for training effects. This study was conducted in the Department clinic where master’s students completed their counseling practicum as part of their degree requirement. Data were collected between Jan 2021 and Jun 2023, over 5 batches of practicum students (20 weeks per round of practicum). To examine counselors’ and clients’ use of emotion words in counseling, eight video-recorded sessions from each of the 64 dyads were transcribed by undergraduate research assistants. The 512 session transcripts generated were processed using the Linguistic Inquiry and Word Count (LIWC) program to identify positive and negative emotion words (Pennebaker et al., 2015). To accommodate the unique linguistic features of Cantonese colloquial speech, we used the Traditional Chinese LIWC (TC-LIWC) dictionary developed in Taiwan (http://cliwc.weebly.com) as a foundation and enhanced its lexicon by incorporating Cantonese expressions extracted from counseling transcripts collected in previous studies. This improved the identification and classification of Cantonese words in the current transcripts. The LIWC was also used to categorize the emotion words into positive and negative emotion words for subsequent analyses. In terms of counseling process and outcome, clients completed measures of session quality and perceived therapist empathy, and counselors completed a measure of session quality, after each counseling session. Objective 2: At the start of practicum training, all counselors completed measures regarding the intensity of their emotional experience, emotional regulation ability, and flexibility in emotional expression. These measures were analyzed in relation to the effects of pre-session lexicon training. The findings for Objectives 1 and 2 were presented at a conference in Krakow, Poland (see attachment in Part C9). Objectives 3: The positive and negative emotion words used by counselors and clients were extracted from session transcripts using the LIWC as described above. Instead of using the number of emotion words in the analyses, we used the proportion of emotion words (i.e., number of emotion words / total number of words spoken in a session) to account for the different speakers’ lengths of speech (clients usually say more than counselors in counseling). In addition, given the nested data structure, where sessions (Level 1) are nested within dyads (Level 2), multilevel modeling was conducted. We examined how the proportion of emotion words used by counselors and by clients predicted therapist empathy and session quality ratings. We also observed the covariation between counselors’ and clients’ proportions of emotion words used in the models (see Figure 1 in publication under review, Part C8). Objective 4: Similarity in counselors’ and clients’ emotion word use were calculated using two metrics: Language Style Matching (LSM) and linguistic style coordination (LSC). LSM scores range from 0 to 1, with a higher value indicating greater similarity of word use between clients and counselors in a session. LSC is directional in nature and assesses level of coordination from one speaker to another in adjacent speaking turns. Specifically, LSC (t to c) refers to the therapist’s use of emotion words in speaking turn N+1 following the client’s use of emotion word use in speaking turn N, and LSC (c to t) refers to the reverse, where the client follows the therapist’s use of emotion words. LSC scores range from -1 to 1; a more positive value indicates greater probability of one following another’s word use (i.e., coordination) and a more negative value indicates greater probability of one not using a word type after the other speaker uses it (i.e., incoordination). LSM and LSC of counselors' and clients' positive and negative emotion words used in sessions were used to predict therapist empathy and session quality ratings (See Figures 2 and 3 in publication under review, Part C8). Objective 5: Objective 5 was achieved using two samples of counseling session transcripts. First, to develop the word list for pre-session emotion lexicon training in this study, we analyzed 160 session transcripts collected from a previous study. The detailed procedure was reported in the mid-term report and is summarized here for ease of reference. Based on client ratings of therapist empathy, sessions were classified as high or low empathy if the empathy ratings were 1 SD above or below the mean, respectively. Using this criterion, 23 sessions were classified as high empathy and 26 as low empathy. From the high-empathy sessions, 124 emotion words used by counselors were identified and included in the training list. Additionally, 190 emotion words used by counselors in both high- and low-empathy sessions were rated by 31 counselors – who were not involved in the transcribed sessions – based on their usefulness for emotional exploration in counseling. The top 116 words from this rating were selected and combined with the initial 124 words to create a 240-word list for the training intervention in this study. Second, we applied the same criteria (1 SD above or below the mean) to classify the 512 sessions collected in the present study as high- or low-empathy sessions. Additionally, sessions were classified as high or low quality based on scores from the Session Evaluation Scale. Using these criteria, 86 sessions were classified as high empathy and 83 as low empathy, while 73 sessions were considered as high quality and 93 as low quality. The emotion words used by counselors and clients were examined in highly rated versus lowly rated sessions, analyzed separately for therapist empathy and session quality. Specifically, we focused on emotion words used exclusively in highly rated sessions, exclusively in lowly rated sessions and those found in both types of sessions. Objective 6: Using the 240-emotion word list described under Objective 5, we created an online resource for counselor training (see attachment under other impact, Part C). In this resource, we summarized the procedure of locating emotion words using counseling transcripts, client ratings, and counselor ratings. We also separated the emotion words by emotional valence (positive or negative or general affect) and source (client-rated or counselor-rated). The online resource is posted on the PI and Co-I (Sarah Luk)’s personal university department webpages for public reference. It has also been shared with trainers in counselling programs across Hong Kong. In addition, 112 trainees (including the 64 participants in this study) from the PI’s institution have been trained using this resource. | ||||||||||||||||||||||||||||
| Summary of objectives addressed: |
|
||||||||||||||||||||||||||||
| Research Outcome | |||||||||||||||||||||||||||||
| Major findings and research outcome: | We did not detect any effect of emotion lexicon training on counselors’ and clients’ emotion word use, clients’ ratings of session quality or therapist empathy, or counselor’s ratings of session quality (Objective 1), even after accounting for counselors’ emotional profiles (Objective 2). These findings are reported at Conference #2 (Part C9). Objectives 3: Counselors’ use of positive and negative emotion words was positively associated with clients’ use of positive and negative emotion words. However, counselors’ use of either did not predict clients’ ratings of session quality or therapist empathy. Interestingly, when clients used more positive emotion words, they rated their counselors as less empathic. Objective 4: Similarity between counselor’s and client’s emotion word use was assessed using linguistic style matching (LSM) and language style coordination (LSC). LSM, which denotes the level of similarity in percentages of emotion word used between counselors and clients, did not predict clients’ ratings of session quality or therapist empathy. LSC, referring to the directional coordination between clients and counselors in a pair of speaking turns, was related to client outcomes: When counselors followed clients’ use of negative emotion words more, clients rated session quality and therapist empathy lower. In contrast, when clients followed counselors’ use of negative emotion words more, clients rated session quality higher. The findings described under Objectives 3 and 4 have been reported in the submitted manuscript (Part C8). Objective 5: Based on the analysis of transcripts from two samples (160 and 512 sessions), we observed that 61-72% of the emotion words used by counselors in highly rated sessions (i.e., on therapist empathy or session quality) were also used by counselors in lowly rated sessions. Furthermore, 74-78% of the emotion words used by clients in highly rated sessions were also used by clients in lowly rated session. These results suggest that there are more common than unique emotion words across the two types of sessions. Further analysis of emotion words is planned and described under 6.2. Objective 6: The 240-emotion word list used in the training study is now published as a training resource on the PI and Co-I (Sarah Luk)’s department profile website (see other impact, Part C). It has also been shared with program leaders of five counseling master’s programs across Hong Kong for training purposes. | ||||||||||||||||||||||||||||
| Potential for further development of the research and the proposed course of action: |
While this study examined positive and negative emotion words in separate models, recent studies on emodiversity in psychotherapy contexts (Alfi-Yogev et al., 2024; Zümbül et al., under review) suggest that our data can be further analyzed in terms of the variety of emotion words present in sessions and its relationship with therapy process and outcome. In particular, individuals who experience a broader range of emotions tend to have better mental health (Quoidbach et al., 2014). Therefore, discussing a greater variety of emotions in counseling may be associated with improved counseling processes and outcome. Additional analyses can be conducted on the present data by excluding emotion synonyms or repeated use of the same emotion words, allowing us to examine the impact of emodiversity in therapist-client conversations on counseling processes. The lack of significant effect of training may be related to the brevity of word presentation right before counseling sessions. Future training interventions could incorporate longer exposure periods, role-plays, and group discussions to enhance the learning and application of emotion vocabulary in counseling. For example, the current list of emotion words can be used as teaching materials in counseling skills courses, and its impact on student learning can be investigated. | ||||||||||||||||||||||||||||
| Layman's Summary of Completion Report: | This study explored whether training counselors to use emotion words would impact therapy sessions. The training did not significantly change how counselors or clients used emotion words, nor did it improve clients’ ratings of session quality or therapist empathy, even when considering counselors’ emotional styles. Although counselors’ emotion word use did not influence clients’ perceptions, counselors and clients tended to mirror each other’s use of emotion words. Nevertheless, simple matching of emotion word use did not predict client ratings. The timing of emotional word use mattered: When clients followed counselors’ negative emotion words, clients rated session quality higher, suggesting they appreciated counselors leading the emotional exploration. Conversely, when counselors followed clients’ negative emotion words, clients rated session and empathy lower, possibly perceiving the counselor as dwelling on distress without providing support. An emotion word list was developed from analyzing counseling transcripts and client/observer ratings. This list is now shared with counseling programs in Hong Kong for training purposes. Future research could explore the impact of using a wider variety of emotion words in sessions and test whether longer and more interactive training could better enhance counselors’ use emotion vocabulary in therapy. | ||||||||||||||||||||||||||||
| Research Output | |||||||||||||||||||||||||||||
| Peer-reviewed journal publication(s) arising directly from this research project : (* denotes the corresponding author) |
|
||||||||||||||||||||||||||||
| Recognized international conference(s) in which paper(s) related to this research project was/were delivered : |
|
||||||||||||||||||||||||||||
| Other impact (e.g. award of patents or prizes, collaboration with other research institutions, technology transfer, etc.): |
|||||||||||||||||||||||||||||
| SCREEN ID: SCRRM00542 |