應用自動化反應技術評量同理心

Apply automatic evaluative reaction processes to assess empathy

楊梵妤(Yang Fan-Yu)

指導教授 ： 林正昌

國立臺灣師範大學/教育學院/教育心理與輔導學系/博士(2020年)

https://doi.org/10.6345/NTNU202001610

摘要

本研究的目的是在發展更適合、有效的同理心 (本研究是指同理關懷) 測量方法。研究者以兩種自動化反應測量進行設計：點偵測與AMP。一共進行了兩個研究以檢驗其信效度。在研究一中，先招募 54 名大學生參與預備性研究，選出作為正式研究中使用的實驗促發圖片。於正式研究中，再招募 50 名心理相關科系的大學生，參與者必須完成兩種同理心自動化測量實驗、人際反應量表(包含EC與 PT 分量表)、新式成人利社會行為量表 (NSAP) 、與三種利社會行為測量 (捐款次數、捐款金額、抄寫次數)。結果發現兩種自動化測量有聚斂效度，並與自陳測量間有區辨效度，但只有同理心 AMP 可以預測抄寫次數。研究二中，一共招募了 135名大學與研究生參與預備性研究，以選出做為正式研究中的文字與圖片材料。正式研究參與者共 47 人，大部分為心理相關科系的大學生。研究二將同理心點偵測將材料換成詞語，而同理心 AMP 則增加了不符合同理的霸凌圖片，參與者須完成自動化測量、自陳測量、與行為測量。結果發現不同類型的測量間有區辨效度，但仍然只有同理心 AMP 能預測抄寫次數。總結兩個研究的結果如下：(一) 同理心點偵測具有信度但不具效度，且無法預測利社會行為。(二) 同理心AMP 有良好的信度，且在典型的利社會行為 (抄寫次數) 上，具有同時效度與預測效度。(三) 同理心 AMP 具有聚斂效度與區辨效度。(四) 同理心 AMP 與EC 沒有相關，可能是因為測驗結構的差異，一種是自動化反應測量，另一種是外顯自陳五點量表，兩者結構完全不相同。而另一個更重要的因素，是同理心層次的不同。 EC 量表所測得的，是外顯的、由個人意向控制的同理關懷；而同理心 AMP 所測得的，是自動化引發的、無法由意向控制的同理關懷。最後，本研究提出了同理關懷的自動化特性，在未來研究的建議上，參與者的組成背景對本研究結果可能造成影響，後續研究可招募更多元的參與者以進行探究。

關鍵字

同理心 ； 自動化反應測量 ； 情感錯誤歸因程序 ； 點偵測

並列摘要

The purpose of this study is to develop a more appropriate and effective measurement for empathy (the “study” here refers to empathic concern). The researcher adopts two types of automatic measurements for the research design: dot-probe task and AMP. A total of two researches were conducted to test the reliability and validity. In research 1, a total of 54 undergraduate students were recruited to participate in a preparatory research to screen out experimental priming pictures to be used in the formal study. In the formal study, 50 undergraduate students from the departments related to the field of psychology were recruited, and they were asked to complete two types of automatic measurement experiments for empathy, Interpersonal Reactivity Index (IRI), New Scale for Adult’s Prosocialness (NSAP) and other three types of prosocial behavior measurements (number of donations, amount of donations, and number of copies). The results show that the two automatic reaction measurement have convergent validity and also they have discriminant validity with self-report inventory, but the number of copies can only be predicted by the empathy AMP. In research 2, a total of 135 of undergraduate and graduate students were recruited to participate in a preparatory study to screen out text and picture materials to be used in the formal study. The 47 participants in the formal study were mostly undergraduate students from the departments related to the field of psychology. Research 2 replaced materials used in the dot-probe task for empathy with texts, and a variety of bullying images that are not conforming to empathy were added to the empathy AMP. The participants were required to complete the automatic reaction measurement, self-report inventory, and behavior measurement. The results show that the different measurements have discriminant validity, but the number of copies can only be predicted by the empathy AMP. The results of the two researches are summarized as follows: (1) Dot-probe task for empathy has reliability but not validity, and it does not predict prosocial behaviors. (2) The empathy AMP displays good reliability, and also offers concurrent validity and predictive validity in terms of the typical prosocial behaviors (number of copies). (3) The empathy AMP has convergent validity and discriminant validity. (4) The empathy AMP and EC are not correlated, which may be due to the difference in test structure, as the structures of the automatic response measurement and explicit self-report five-point Likert scale are totally different. Another more important factor is the different levels of empathy. The EC scale measures the empathic concern that is explicit and controlled by personal intentions, while the empathy AMP measures empathy that is triggered by automation and not controlled by intentions. Finally, this study proposes the automatic characteristics of empathic concern. With respect to the recommendations for future studies, the background of participants may also affect the results of this research, so that it is desirable to have participants with diverse backgrounds for further studies.

並列關鍵字

Empathy ； Automatic Evaluative reaction ； AMP ； Dot Probe

參考文獻

一、中文部分

林思賢 (2009)。情感錯誤歸因程序的歷程分離：激發後校正模式 (未出版之博士論文)。國立中正大學，嘉義縣。

林苡彤、程景琳 (2010)。國中生關係攻擊加害者與受害者之規範信念、同理心與因應策略。台東大學教育學報，21 (2)，1-28。

郭慧婷 (2019)。大學生社會連結感對利社會行為的影響：以同理心作為中介變項 (未出版之碩士論文)。國立臺灣師範大學，台北市。

蔡孟寧、王倫婷、林烘煜 (2015)。情感錯誤歸因程序的自動化反應態度測量檢視。中華心理學刊，57，261-280。 [Tsai, M. N., Wang, L. T., & Lin, H. Y. (2015). Affect misattribution procedure as an automatic evaluative reaction measurement.Chinese Journal of Psychology, 57, 261-280. DOI: 10.6129/ CJP.20150202]