Lo(u)st in Therapy: Development and Psychometric Evaluation of the Therapists’ Attitude toward Sexual and Erotic Feelings Scale (TASEF)

Abstract This study aimed to develop the Therapists’ Attitude toward Sexual and Erotic Feelings (TASEF) scale for assessing therapists’ attitudes and responses toward sexual and erotic feelings in therapy. A total of 507 therapists participated in two separate studies. Study 1 focused on finalizing the format of the TASEF and evaluating its construct validity. Through Exploratory Graph Analysis, four factors were identified within the TASEF questionnaire: Gratification, which captures the positive affect experienced by therapists in response to a patient’s sexual attraction toward them; Fear, which represents therapists’ apprehension or discomfort when encountering sexual feelings from the patient; Threat, which reflects the perception of risk or danger that erotic transference poses to the patient, therapist, and the therapeutic process; and Aversion, which encompasses the therapist’s discomfort or aversion toward the patient’s sexual feelings. Study 2 aimed to confirm the structure of the TASEF questionnaire and assess its overall validity. The results showed that the TASEF demonstrated adequate overall validity, with higher convergence validity observed for the Fear, Aversion, and Threat clusters compared to the Gratification cluster. The TASEF enables therapists to understand and manage their emotional reactions to sexual feelings in therapy, maintaining standards of care.


Introduction
In psychoanalytic theory, sexual and erotic feelings are commonly understood and conceptualized as manifestations of erotic transference and countertransference within the therapeutic relationship.Recognizing the significance of exploring therapists' broader attitudes toward these erotic themes is crucial for gaining a comprehensive understanding of the dynamics at play in therapeutic settings.Erotic transference, also known as sexual transference, is a phenomenon that occurs in the therapeutic relationship where the patient develops feelings of sexual attraction toward their therapist (Blum, 1973).The term erotic transference is psychoanalytic in origin, deriving from Freud and Strachey's (1915) paper "Observations on Transference Love."Erotic transference is defined as any of the patient's fantasies about the therapist that contain elements that are primarily romantic, intimate, sensual, or sexual (Book, 1995).Erotic countertransference refers to instances where the therapist develops feelings of sexual attraction toward the patient (Freud & Strachey, 1915).This phenomenon is considered a normal part of the therapeutic process and has been studied extensively in the field of psychoanalysis (Book, 1995;Celenza, 2010;Krausz, 2016;Lijtmaer, 2004;Mann, 1997;Russ, 1993;Welles & Wrye, 2013).According to psychoanalytic theory, erotic transference is a defense mechanism used by patients to prevent them from remembering painful memories from the past.It is also seen as a way for the patient to replay early interactions and feelings of love and desire in the therapeutic alliance (Welles & Wrye, 2013).If the capacity for love was damaged in early childhood experiences, patients may attend their sessions for closeness to the therapist, with the hope that the therapist will reciprocate their love (Rachman, Kennedy, & Yard, 2009).
Research has shown that when handled properly, sexual feelings in therapy can be a positive force in the therapeutic process, helping patients to understand and work through underlying issues related to attachment, sexuality, and intimacy (Little, 2018).To ensure the sensitive and professional handling of sexual and erotic feelings in therapy, therapists should receive appropriate training and develop expertise in effectively managing and navigating these aspects within the therapeutic context.Therapists often lack both knowledge on erotic transference and clear guidelines for effectively addressing it, leading to a lack of confidence in handling erotic issues and setting clear boundaries (Barnewall, 2016;Colom-Timlin, 2014).
The presence of the therapist's sexual feelings in therapy, commonly referred to as erotic countertransference in psychoanalytic theory, can introduce complexities and challenges within the therapeutic relationship, which can have the potential to create ethical and boundary issues that have the potential to harm the patient and compromise the integrity of the therapeutic process.Therefore, it is important to help therapists evaluate their own reactions to sexual feelings and identify any potential issues related to erotic countertransference.By examining the therapist's own experience of sexual feelings toward the patient (erotic countertransference), it is possible to identify potential obstacles to effective therapy and take appropriate steps to address them.This can help to ensure that the therapist remains focused on the patient's needs and maintains appropriate professional boundaries, while also promoting a safe and supportive therapeutic environment.
The phenomenon of erotic countertransference has been discussed significantly less than erotic transference.Research has shown that erotic countertransference, measured by sexual attraction and erotic feelings toward the patient, is a common experience among therapists.Experiencing sexual feelings toward the patient can be triggered by various factors such as unresolved personal issues, lack of boundaries, or specific characteristics of the patient (Mann, 1997).Sexual attraction toward a patient has been reported by 95% of male therapists and 76% of female therapists (Pope, Keith-Spiegel, & Tabachnick, 1986).In a more recent study, seven out of ten (70.6%) therapists reported they found a patient sexually attractive, and almost a quarter (22.8%) fantasized about a romantic relationship with a patient (Vesentini et al., 2022).Pope and Tabachnick (1993) noted differential treatment of female clients, with therapists more likely to perceive them as physically attractive.
Experiencing erotic and sexual feelings toward a patient can be particularly challenging for the therapist to manage as it has the potential to compromise the therapeutic relationship and harm the patient.Therefore, it is important for the therapist to be aware of these feelings, and to handle it in a sensitive and professional manner.In qualitative studies that have explored this matter through interviews with therapists, therapists have described feeling angry at themselves for being sexually attracted or aroused in therapy, as well as confused by their arousal (Rodgers, 2011).Additionally, some therapists tend to hold moralistic attitudes, considering any form of sexual dynamics within therapy to be ethically wrong (Luca, 2018;Nickell, Hecker, Ray, & Bercik, 1995).However, it should be noted, that the occurrence of therapists acting on their sexual feelings instead of seeking supervision to address countertransference is relatively infrequent and does not reflect the predominant behavior of therapists (Gabbard, 1994(Gabbard, , 1996)).
Feelings of shame and guilt, and moralistic attitudes, may lead therapists to embrace more defensive and avoidant strategies such as wearing a wedding ring, ignoring erotic transference, or telling patients that sexual involvement is prohibited (Luca, 2018;Pope et al., 1986;Rodgers, 2011;Spilly, 2008).The defensive and avoidant approach toward erotic and sexual feeling that arise in therapy can also be attributed to the absence of discussing this issue in core training.Therapists often lack knowledge and guidelines for effectively managing erotic and sexual feelings that arise in therapy.According to a study involving 107 therapists (Barnewall, 2016), they find it difficult to set boundaries in handling these feelings.As a result, therapists lack confidence in handling erotic issues and in setting clear boundaries in an attempt to managing sexual and erotic feelings (Barnewall, 2016).
Although therapists acknowledge the value of exploring erotic and sexual feeling within the therapeutic relationship (Rodgers, 2011), they often tend to deny or dismiss these feelings because they are afraid of frightening the patient or simply feel uncomfortable in discussing sexual-related issues.In cases where sexual and erotic feelings were reciprocal, therapists were reluctant to raise this issue during supervision and experienced higher levels of anxiety and discomfort (Spilly, 2008).

The need for measuring therapists' attitude toward sexual and erotic feelings
Previous literature has made a significant contribution to estimating the prevalence of erotic countertransference by examining sexual and erotic feelings that emerge in therapists toward their patients (Giovazolias & Davis, 2001;Pope et al., 1986), and in describing the lack of supervision and training in this area (Krausz, 2016;Meritt, 2011;Pope et al., 1986;Sehl, 1998;Vesentini et al., 2022).However, there has been significantly less emphasis placed on identifying and comprehending the diverse responses that therapists may have in relation to their patients and their own sexual and erotic feelings that emerge toward their patients.Thus, the main objective of this study is exploring and understanding the multifaceted responses of therapists toward patients' and their own sexual and erotic feelings in therapy.Considering the limited availability of measures in assessing this phenomenon, the study aims to develop a scale, referred to as the therapists' attitude toward sexual and erotic feelings scale, specifically designed to evaluate therapists' attitudes and responses toward sexual and erotic feelings in therapy.To accomplish this, two separate studies were conducted.Study 1 was used to finalize the format of the TASEF questionnaire and evaluate its construct.In Study 2, the researchers examined the relationship between attitudes and responses toward sexual and erotic feelings in therapy (measured by the TASEF scale) and attachment orientation, differentiation of self, and confidence in discussing sexuality.
The selection of attachment orientation, differentiation of self, and confidence in discussing sexuality is driven by their relevance to understanding therapists' responses toward sexual and erotic feelings in therapy.Attachment orientation, which refers to individuals' internalized patterns of relating and connecting with others (Bowlby, 1982), can be associated with how therapists perceive and respond to intimate and sexual dynamics in the therapeutic relationship.Differentiation of self includes two interrelated dimensions: the intra-psychic, which reflects one's ability to regulate emotions and intellect (Bowen, 1985), and the interpersonal, which reflects the capacity to regulate closeness and autonomy (Bowen, 1985;Bowen & Kerr, 1988).Thus, therapists with higher levels of differentiation may be better equipped to manage their own emotional reactions and establish appropriate boundaries when confronted with sexual and erotic feelings in therapy.Finally, confidence in discussing sexuality is an important factor to consider as it reflects therapists' comfort, knowledge, and skill in addressing sexual topics within therapeutic settings (Harris & Hays, 2008).
With the aim of achieving practical implications, it is hoped that the TASEF will serve as a valuable measure to assess the therapist's attitudes and responses toward sexual and erotic feelings in therapy.This can be important for normalizing these feelings, and for therapist's self-reflection, which will enable therapists to gain insight into their own emotional reactions to sexual and erotic feelings that arise in therapy; for monitoring progress and changes throughout the therapeutic process; for assessing the prevalence and impact of therapists' attitudes and responses toward sexual and erotic feelings in therapy; and for maintaining standards of care ensuring therapists are providing the best possible care for their patients and that any potential harm to the patient is avoided.

Study 1
Study 1 was conducted to finalize the format of the TASEF questionnaire and evaluate its construct.

Sampling and procedure
To be included in the present study, participants had to (a) have at a minimum a bachelor's degree in the following professions: social work, psychology, psychotherapy, counseling, or couples and sex therapy, and (b) be able to read and answer a survey in Hebrew.The study was approved by the ethics committee of the University of Haifa.Participants did not sign informed consent but received detailed information concerning the study prior to their responses.The survey was conducted between March and May 2021.Participants filled out an online survey using Qualtrics, a web-based survey and data service.Qualtrics servers are protected by high-end firewall systems and promises anonymity.Participants completed the link voluntarily.After the university ethics committee approved the study, an anonymous link using Qualtrics software was distributed via social media.Participants were invited to take part in an anonymous study on erotic transference and countertransference.No incentives were given to participants.

Measures
Background variables included a brief demographic questionnaire that assessed gender, sexual orientation, age, religion, religiosity, relational status, income, self-rated health, profession, education and professional training programs or specialized certification programs, and years of clinical experience.
Responses to sexual and erotic feelings in therapy was assessed using the Therapists' Attitude toward Sexual and Erotic Feelings scale (TASEF).Items were developed based on consultants with clinical experts, and based on gaps in the literature.Toward developing the scale, the authors have conducted a qualitative study to identify different reactions among therapists that were then qualitatively analyzed (Author's own, under review).Based on the thematic analysis, consultation with five clinical experts, and their own clinical experience (all authors are also clinicians), the authors have generated a pool of 24 items.Five clinical experts who specialize in psychotherapy, couples therapy, psychoanalytic therapy and sex therapy, assisted with developing the initial pool of items and suggested another 11 items.The initial pool consisted of 60 items.These experts were also encouraged to offer modifications in the wording of the items developed by the authors, and to make note of any items they felt were missing, or of any redundancies.To assess the clarity of the questionnaire for laypeople, the scale was also presented to three individuals (university students with whom the principal investigators were acquainted via their social networks).These individuals were asked to read the items and to indicate any item whose meaning was unclear.As a result of these procedures, minor modifications were made in the wording of the items, and 2 items were omitted due to redundancy.The final version of the questionnaire included a total of 33 items.Participants were required to rate the extent to which each reaction applied to them, using a scale ranging from 1 (not at all) to 5 (very much).

Data analysis
To examine and optimize the factorial structure of the TASEF, we employed Exploratory Graph Analysis (EGA; Golino et al., 2019) using EGAnet R package-a network psychometrics method that uses undirected network models for the assessment of psychometric properties of questionnaires.EGA was used to verify the number of factors using a graphical lasso (Friedman, Hastie, & Tibshirani, 2008) and the items that are associated with each factor.Network loadings, roughly equivalent to factor loadings, are reported using net.loads(), with suggested general effect size guidelines for network loadings of 0.15 for small, 0.25 for moderate, and 0.35 for large (Christensen & Golino, 2021).The number of factors was corroborated with other traditional methods-parallel analysis (PA), Velicer's minimum average partial (MAP) test, and the comparison data approach (Ruscio & Roche, 2012).The stability of the EGA was appraised using Bootstrap Exploratory Graph Analysis with 5,000 resampling cycles.After the initial EGA, we optimized the selection of items through a series of iterations.We dropped items with network loading lower than 0.15 until a final version of EGA comprised factors with items with greater network loading than 0.15.The structural stability of the final version was reappraised using Bootstrap Exploratory Graph Analysis with 5,000 resampling cycles.We also estimated the stability of each item by the itemStability() function, and corroborated the adequateness of the final form of the questionnaire by a Confirmatory Factor Analysis (CFA) using lavaan Structural Equation Modeling (SEM) R package, with a maximum likelihood estimation with robust standard errors and a mean and variance adjusted test statistic.
The data and materials used in the research can be available upon request.

Participants
A total of 116 participants − 101 women, 15 men.Of them, 101 (86.3%) identified as heterosexual.The mean participant age was 45.84 (SD = 9.68, age range: 28-70).The majority of participants identified as either married or in a relationship (75%), and the rest were either separated or divorced (13.8%) or single (11.2%) and the majority of respondents had children (n = 97, 83.6%).The sample was highly educated, only three people (2.6%) had a bachelor's degree, 99 (84.6%) held a master's degree, and 14 (12.1%) had a PhD.The majority of participants (n = 114, 97.4) were Jewish, and non-religious (82.9%).The majority of participants (61.5%) reported an income higher than the average income in Israel (10,584 NIS).More than half of the sample (n = 63, 53.8%) were social workers, and the rest had a degree in psychology (n = 17, 14.5%), or other therapeutic degrees (n = 36, 31.7%),such as art therapy and family-therapy.The majority of the sample reported undergoing continuing education courses and certifications (following their academic degree), such as psychotherapy (n = 46, 39.3%), couples therapy (n = 38, 32.5%), sex therapy (n = 25, 21.4%), and other certifications (n = 19, 16.2%) such as EMDR, CBT, art therapy, somatic experience (SE) and Post Traumatic Transformation (PTT).The sample consisted of very experienced therapists, with 25.6% reporting 20 and more years of clinical experience, 15.4% reporting 15-20 years of clinical experience, 24.8% reporting 10-15 years of clinical experience, 26.5% reporting 5-10 years of clinical experience, and only 6.8% reporting less than five years of clinical experience.

Exploratory graph analysis
The initial EGA of the TASEF indicated that the factorial structure comprised four factors: "gratification" consisting of 9 items, "fear" consisting of 11 items, "aversion" consisting of 7 items, and "threat" comprising 6 items.A 4-factor solution was corroborated by two additional analyses-parallel analysis (eigenvalues of 4.47, 3.43, 1.73, and 1.22 for the 4 factors), and the Comparison Data.Conversely, Velicer's MAP (squared, .022,and 4th power: .0015)suggested that only three factors are needed to retain.When estimating the stability of the EGA by bootstrapping with 5,000 resampling cycles, the analysis indicated low stability: SE = .78,with CI for the number of factors ranging from 2.46 to 5.53; the 4-factor solution was only prevalent in 57.94% of the bootstrap samples, with 13.11% producing a 3-factor solution, 22.98% producing a 5-factor, and 4.86%-a 6-factor solution (with less than 1% producing 1, 7, and 8-factor solutions).We, therefore, optimized the questionnaire by dropping items with low network loading.A total of 14 items were dropped.The final 19-item questionnaire had high structural stability: SE = .38,with CI for the number of factors ranging from 3.24 to 4.75 and a prevalence of the 4-factor solution of 83.26% of the bootstrap samples (with 15.92% producing a 3-factor solution, and 7. 8% producing a 5-factor).The initial and final EGAs are presented in Figure 1, and network loadings are in Table 1.In addition, all items had stability higher than 75% (see Figure 2).A confirmatory factor analysis that was used to corroborate the EGA solution verified the factorial structure, χ 2 (67.40) = 135.55,p < .01,CFI = .98,TLI = .96,RMSEA = .069(90% confidence interval [CI] of .059,.079),SRMR = .067(see Figure 3).

Study 2
The aim of study 2 is to validate and confirm the structure of the TASEF and assess its overall validity of the scale and the factors (Fear, Aversion, Threat, and Gratification).To assess the discriminant validity of the TASEF, we used attachment orientation, differentiation of self, and confidence in discussing sexuality.We hypothesize that the scale will demonstrate a valid and reliable structure, consisting of four distinct factors: Fear, Aversion, Threat, and Gratification.To assess the discriminant validity of the TASEF, we predict that there will be statistically significant, yet weak, associations with three additional constructs: attachment orientation, differentiation of self, and confidence in discussing sexuality.These selected constructs are expected to demonstrate limited shared variance with the TASEF factors, allowing for precision in differentiating between them.

Sampling and procedure
The same inclusion criteria, sampling, and procedure were applied as previously described in Study 1. Study 2 was conducted between April and November 2022.

Measures
Background variables were identical to those described in Study 1. Therapists' attitude toward sexual and erotic feelings in therapy was assessed using the Therapists' Attitude toward Sexual and Erotic Feelings (TASEF), previously described in Study 1.
Differentiation of Self was assessed using the using the Differentiation of Self Inventory-Short Form (DSI-R; Sloan & van Dierendonck, 2016) which is a shortened version of the DSI scale (Skowron & Friendlander, 1998).The DSI-R is a 20-item version of the Differentiation of Self Inventory-Revised (DSI-R; (Skowron & Schmitt, 2003).The DSI-R measures the degree to which an individual can balance emotional with intellectual functioning and closeness with independence in relationships with others by examining four factors: I-position (e.g., "No matter what happens in my life, Iknow that I'll never lose my sense of who I am"), emotional reactivity (e.g., "I'm very sensitive to being hurt by others"), emotional cutoff (e.g., "When one of my relationships becomes very intense, I feel the urge to run away from it"), and fusion with others (e.g., "I often agree with others just to appease them").Emotional cutoff and emotional reactivity represent imbalanced self-differentiation, while fusion-with-others refers to prioritizing others over oneself.Balanced self-differentiation involves developing a separate identity and flexible boundaries.Fused individuals avoid separation, while cutoff individuals perceive closeness as threatening.Similar to the original DSI, it is based on a 6-point Likert-type scale, ranging from 1 (not at all true of me), to 6 (very true of me).The Cronbach's alpha in the present study (total score) was good (α = 0.83).Attachment orientation was assessed using the Experiences in Close Relationships Scale-Short form (ECR-S) (Wei, Russell, Mallinckrodt, & Vogel, 2007) which is a 12-item scale designed to measure adult attachment insecurities in close relationships.The measure has two subscales: attachment anxiety, representing a person's need for approval and anxiety about rejection by others (e.g., "I need a lot of reassurance that I am loved by my partner"), and attachment avoidance, indicating an individual's fear of intimacy with others and hesitance to self-disclose (e.g., "I want to get close to my partner, but I keep pulling back").Respondents use a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree) to indicate their level of agreement with statements about their romantic relationships.The score for each subscale was calculated as an average of the score of each statement.An individual with low levels of attachment anxiety and attachment avoidance was considered to have secure attachment.In the current sample, Cronbach's alpha was acceptable for both the attachment anxiety subscale (α = .67),and the attachment avoidance subscale (α = .60).
Comfort in discussing sexuality in therapy was assessed using the comfort with discuss patient's sexuality scale (Harris & Hays, 2008) which is a 10-item scale designed to measure the level of comfort the therapist has with discussing sexual-related issues in therapy.Therapists were asked to rate their level of agreement on a scale from 1 (strongly disagree) to 7 (strongly agree) how open and confident they feel when discussing sexual issues in therapy (e.g., I encourage patients to explore their own sexual issues, I encourage patients to explore their own sexual values).Score was averaged, with a higher score indicating more comfort in discussing sexual issues in therapy.In the current sample, Cronbach's alpha for the scale was good (α = .85).

Data analysis
We began by reexamining the structural validity of the 19-item TASEF by Confirmatory Factor Analysis (CFA) using lavaan Structural Equation Modeling (SEM) R package, with a maximum likelihood estimation with robust standard errors and a mean-and variance-adjusted test statistic.Next, we adopted the Multi-trait Multimethod analysis for assessing the convergence and discriminant validity of the 19-item TASEF.The analysis was conducted within the psy package with the mtmm() function.The Multi-trait Multimethod analysis examines the within-cluster and between-cluster correlation such that the unit of analysis is a single item.Good psychometric indices are high within-cluster correlations (e.g., high correlations between all items of the gratification cluster of the TASEF), moderate between-cluster correlations between items from similar theoretical constructs (i.e., convergence validity), and weak between-cluster correlations between items from unrelated theoretical constructs (i.e., discriminant validity).In the current study, the convergence validity within the Multi-trait Multimethod analysis was estimated between different constructs of the TASEF (i.e., between-cluster correlations of the gratification, fear, aversion, and threat clusters); the discriminant analysis was appraised by assessing the between-cluster correlations between items of the TASEF and items of differentiation, attachment avoidance and anxiety, and sexual confidence.In the final section of results, we examined the associations between the clusters of the 19-item TASEF and the following background measures-gender, marital status (married, unmarried), sexual orientation (heterosexual, non-heterosexual), sex therapist (yes, no), religiosity (secular, religious), age, years of clinical experience, income, and self-rated health.The associations with the categorical background measures were based on Welch's independent samples t-test followed by Hedges's g and 95% confidence intervals, and the associations with the quantitative measures on Pearson correlations with 95% confidence intervals.

Confirmatory factor analysis
The initial model had an unsatisfactory fit to the observed data, χ 2 (146) = 581.09,p < .01,CFI = .83,TLI = .80,RMSEA = .086(90% confidence interval [CI] of .079,.094),SRMR = .076.Using the model indices, we identified five within-cluster covariates between error terms that are needed to increased model's fit to satisfactory levels: item 11 ("It is terrifying for me when I have sexual fantasies on a patient") and item 17 ("Some patients' descriptions of their sexual experiences can make me feel extremely uncomfortable"); item 2 ("I enjoy the thought that I am attractive in the eyes of some of my patients") and item 4 ("I sometimes experience sexual tension in response to various content raised by my patients"); item 16 ("Some patients' descriptions of their sexual experiences can make me feel extremely uncomfortable") and item 5 ("It is important to me to be seen as attractive in the eyes of some of my patients"); item 5 ("It is important to me to be seen as attractive in the eyes of some of my patients") and item 1 ("I enjoy being attractive to my patients"); item 1 ("I enjoy being attractive to my patients") and item 16 ("Some patients' descriptions of their sexual experiences can make me feel extremely uncomfortable"); and finally item 5 ("It is important to me to be seen as attractive in the eyes of some of my patients") and item 8 ("I'm scared a patient will fall in love with me").These covariates are theoretically logical, and by adding these covariates to the CFA model, it had an excellent fit to the observed data, χ 2 (140) = 268.53,p < .01,CFI = .95,TLI = .94,RMSEA = .048(90% confidence interval [CI] of .039,.057),SRMR = .06.In other words, the factorial structure of the 19-item TASEF was confirmed.

Multi-trait multimethod analysis for appraising convergence and discriminant validity
The results of the Multi-trait Multimethod analysis are summarized in Figure 4 (factor level) and 5 (item level) and Table 2.The analysis revealed that the within-cluster correlations of all four clusters of the TASEF were high and ranged between .49and .76.The convergence validity of the fear, aversion, and threat clusters was adequate (.20-.28); the convergence validity of the gratification cluster was low in comparison (.10).However, all the convergence validities were higher than the discriminant validities signifying adequate overall validity of the TASEF.Specifically, the mean correlations between the TASEF clusters and differentiation, attachment avoidance and anxiety, and sexual confidence were r = .07,r = .09,r = .09,and r = −0.06,respectively.

Associations with background measures
The results are summarized in Tables 3 (categorical measures) and 4 (quantitative measures).The findings revealed that women exhibited a greater aversion to sexual and erotic feelings in  therapy compared to men.Similarly, non-heterosexual individuals showed higher aversion toward sexual and erotic feelings in therapy compared to heterosexual individuals.The most pronounced differences were between sex therapists and other people-sex therapists had significantly higher gratification and lower fear and threat in relation to sexual and erotic feelings in therapy.In addition, the analyses revealed that the higher participants' fear and aversion in relation to sexual and erotic feelings in therapy, the younger, less experience, and lower income they had.Higher income was also associated with higher gratification and lower threat in relation to sexual and erotic feelings in therapy.

Discussion
In this study, we aimed to gain a deeper understanding of the reactions of therapists to sexual and erotic feelings in therapy by developing the TASEF.The limited measures available to assess this phenomenon highlight the need for a standardized and objective tool to evaluate therapists' reactions to sexual and erotic feelings in therapy.The current findings yielded four TASEF factors: Gratification, fear, aversion, and threat.Gratification refers to the therapist's sense of pleasure or excitement in response to the patient's sexual and erotic feelings toward them.This has previously been discussed as the "guilty pleasure" of the therapist (Celenza, 2010).Sexual and erotic feelings in therapy can result in some level of gratification for the therapist by fulfilling their attachment and narcissistic personal needs, making them feel loved, attractive, desired, wanted, and valued.However, gratification may also be a significant ethical and therapeutic concern, as it may lead to the violation of professional boundaries and exploitation of the patient.This is especially true if the therapist has high narcissistic needs (Maroda, 2002), or unmet attachment needs that are projected onto the patients and can increase the risk of using the therapeutic relationship to fulfill those needs, rather than focusing on the patient's needs.
Fear refers to the therapist's sense of apprehension or unease when faced with the patient's sexual feelings toward them, or when experiencing sexual feelings toward the patient.Fear can stem from the risk of crossing professional boundaries, and being tempted to reciprocate the patient's feelings (Gabbard, 1994).Fear can also represent split off or dissociated from erotic feelings (through anxiety; (Little, 2018).It is also possible that fear arises from not knowing how to handle sexual and erotic feelings in therapy.This corresponds with studies examined therapists' experiences and have revealed that many of them feel ill-equipped and insufficiently informed about how to handle sexual and erotic feelings in therapy (Begun, 2011;Sehl, 1998;Spilly, 2008).
The fear of being judged is also a concern for many therapists, and they may worry that their colleagues or patients will view them as incompetent or unprofessional if they are not able to handle the situation correctly.Therapists may also experience fears related to their own attachment issues.This can include a fear of being loved or a difficulty connecting with others due to past relationship trauma.Therapists may be afraid of being seduced or manipulated by their patients, which can create feelings of mistrust and anxiety.Finally, a more psychodynamic perspective offered by Gabbard (1994), suggests that the anxiety of the therapist is generated by concerns about the loss of their identity as a separate person.
Threat refers to the perception of danger or risk posed by sexual and erotic feelings in therapy, including potential harm to the patient, the therapist's professional standing, or the therapeutic process.The reasons that sexual and erotic feelings in therapy can be threatening to a therapist are varied.For one, these feelings can be threatening to the therapist because it may challenge their professional boundaries and lead to a sense of diminished professional esteem.They may feel as if they did something unprofessional on their side that led to these feelings among the patient.In addition, opening up about this issue with the patient can be a delicate process, as it may lead the patient to feel rejected or ashamed (Barnewall, 2016).This may hurt the therapeutic alliance, and the therapeutic process.Finally, sexual and erotic feelings in therapy can also be perceived as threatening to the progress the patient has mad in therapy, as these feelings can reenact early traumas of rejection and feelings of being unloved, and any discussion about this erotic transference can be unsafe and even harmful for the patients, as it can lead to the reliving of the trauma (Davies, 2001).
Aversion refers to the therapist's feelings of discomfort or repulsion in response to the patient's sexual and erotic feelings toward them.This reaction may stem from a variety of factors, including personal values or beliefs, discomfort with discussing sexual matters, or a sense of unpreparedness for managing these feelings, ethical concerns, or discomfort with intimacy.It may also be a natural and understandable response to unwanted or inappropriate sexual or romantic feelings expressed by the patient.Finally, the therapist may feel uncomfortable or repulsed by the patient's feelings because they trigger the therapist's own unresolved issues or traumas around sexuality and intimacy.The therapist may be projecting their own fears, anxieties, or discomfort with intimacy onto the patient's feelings.
Regarding discriminant validity, our findings indicate that the TASEF is indeed distinct from other constructs measured in the study, namely differentiation, attachment avoidance and anxiety, and confidence in discussing sexual content in therapy.This suggests that the TASEF primarily captures unique aspects of gratification, fear, aversion, and threat, signifying its specificity as a robust psychometric instrument.However, the weak correlations between TASEF and these constructs is consistent with our theoretical expectations.Specifically, therapists who reported higher levels of anxious and avoidant attachment orientations, lower levels of differentiation of self, and less confidence in discussing sexuality tended to score differently in their attitudes toward sexual and erotic feelings in therapy, as measured by the TASEF.This highlights the nuanced interplay between the TASEF and these related constructs, reaffirming that the scale is specifically designed to capture unique aspects of gratification, fear, aversion, and threat, while being associated, to some extent, with differentiation, attachment styles, and comfort in discussing sexuality.
We found that women had a higher aversion in relation to sexual and erotic feelings in therapy than men.While this experience is not exclusive to women therapist, the way in which they undergo it may be influenced by their gender.The breach of bodily boundaries is a cause for profound emotional sensitivity, particularly when there is a tolerance for such violations, which highlights the psychological and physical intrusion brought about by the ability to give birth and breastfeed (Russ, 1993).The aversion women therapists experience when confronted with their own sexual and erotic feelings toward a patient may be attributed to the socialization and cultural upbringing of women (Kelly, 2014) in which their sexual feelings are a source of shame, and also hold a greater sense of responsibility (victim blaming) for these feelings if they appear in the therapeutic relationships.Women therapists may also be more likely to experience sexual and erotic feelings in therapy toward a patient because of their own experiences of gendered power dynamics, sexual trauma, or other factors that can complicate the therapeutic relationship.
We also found that therapists who identify as LGBTQ had higher gratification as a response to sexual and erotic feelings in therapy than heterosexual therapists.Therapists who identify as LGBTQ may have higher gratification in response to sexual and erotic feelings in therapy due to two possible factors.Firstly, individuals who identify as LGBTQ may have dealt with issues related to sexuality from a young age and may, therefore, be more sensitive to the topic (Baumann, Ryu, & Harney, 2020).Consequently, when patients express romantic or sexual feelings toward them, they may feel a greater sense of understanding and empathy, which can contribute to feelings of joy or satisfaction.Secondly, therapists who identify as LGBTQ may feel more validated and accepted in their own sexuality when their patients express romantic or sexual feelings toward them.This validation and acceptance can be especially significant as sexual minorities often experience high levels of stigmatization, stigma, and discrimination from society, leading to feelings of invalidation and marginalization (Meyer, 2003).
The most pronounced differences were between sex therapists and other people-sex therapists had significantly higher gratification and lower fear and threat in relation to sexual and erotic feelings in therapy.This finding may be explained by two possible reasons.Firstly, sex therapists receive specialized training in working with clients around issues related to sexuality and intimacy, which may make them feel more confident and competent in managing countertransference reactions and creating a safe and supportive therapeutic environment.Additionally, sex therapists may be more familiar with sexual material, including fantasies, desires, and behaviors, which can make it easier for them to recognize and address sexual and erotic feelings in therapy (Meritt, 2011).As for gratification in relation to sexual and erotic feelings in therapy, it could be argued that sex therapists' gratification in response to sexual and erotic feelings in therapy may be due to their inherent comfort and openness around issues related to sexuality.It is possible that sex therapists may have chosen this field of therapy because they feel drawn to the subject matter and find gratification in discussing these sensitive topics and passion for understanding and helping individuals with sexual concerns.Their interest and expertise in the subject matter can lead to a sense of gratification when they can use their knowledge to assist their clients in resolving sexual and erotic issues.Additionally, they might sublimate their own feelings or personal biases to focus on the well-being of their clients, and successfully doing so can also lead to feelings of satisfaction and accomplishment.Insufficient training in psychotherapy among sex therapists could potentially lead to an increased risk of experiencing gratification from sexual and erotic feelings in therapy.This is because without a strong foundation in psychotherapy and the necessary skills to manage countertransference reactions, therapists may be more susceptible to experiencing sexual and erotic feelings in therapy and struggling to maintain appropriate boundaries with their clients.
Our analyses revealed that younger and less experienced therapists reported higher participants' fear and aversion.Younger individuals and those with less experience may be less comfortable discussing topics related to sexuality and intimacy.As a result, they may be more likely to feel uncomfortable or afraid when confronted with sexual and erotic feelings in therapy.
Higher income was also associated with higher gratification and lower threat in relation to sexual and erotic feelings in therapy.This finding was surprising to us and requires further investigation.It is possible that individuals with higher income may occupy positions of power and privilege in their personal and professional lives, which can create a sense of entitlement and influence in relationships, including in therapeutic relationships.This power dynamic can lead to a higher likelihood of gratification from sexual and erotic feelings in therapy and feeling less threatened by the experience.

Theoretical and clinical contribution
The development of the TASEF scale represents a significant theoretical contribution to the field of psychotherapy.The limited measures available to assess erotic countertransference make it difficult to fully understand this phenomenon and develop effective interventions to address it.The TASEF provides a standardized and objective tool to measure the therapist's reactions to sexual and erotic feelings in therapy.This has important implications for therapist self-reflection, as it can help therapists gain insight into their own emotional reactions and work to manage them in a way that does not harm the patient or the therapeutic relationship.Additionally, the TASEF can be used to monitor progress and changes throughout the therapeutic process, assess the prevalence and impact of sexual and erotic feelings in therapy among therapists, and maintain standards of care to ensure that therapists are providing the best possible care for their patients while avoiding potential harm.Overall, the development of the TASEF represents a significant step forward in understanding and managing sexual and erotic feelings in therapy.
The experience of sexual and erotic feelings in therapy is a complex and challenging issue that requires careful management by the therapist.It is important for therapists to approach this issue with empathy, compassion, and a commitment to maintaining appropriate professional boundaries.The therapist's professional responsibility is to identify the sexual and erotic feelings in therapy and not avoid its existence, as discussing an erotic transference can be helpful for the therapeutic process.This discussion allows patients to learn a great deal about themselves and their relationships with others.Additionally, acknowledging the therapist's own biases and managing them in a way that does not harm the patient, or the therapeutic relationship is important.It is also crucial for therapists to be aware of the potential for an erotic countertransference and to interrupt or stop patients from expressing feelings of love for their parents, as this can lead to dissociation or emotional overload.By navigating these dynamics effectively, therapists can create a safe and supportive environment that promotes growth and healing for both the patient and the therapist.

Limitations and future studies
It is important to consider the limitations of the present study when interpreting its results.One major limitation is that the study relied on self-report measures, which could be influenced by response biases and shared method variances.In future studies, the validity of the TASEF should be examined among diverse samples of therapists and health professionals to ensure that the results are generalizable.It should be emphasized that there is currently no direct evidence to support the claims regarding the reasons why sex therapists and therapists who identify as LGBTQ may experience higher levels of gratification from sexual and erotic feelings in therapy.Further research is needed to fully understand the complex factors underlying these dynamics and to determine the extent to which factors such as specialized training, personal characteristics, or other factors may play a role.Further scientific investigation is necessary to examine the factors of the TASEF in relation to specific content areas of therapy that may present opportunities for aversion or fear countertransference.These areas include topics such as anal and oral sex, non-monogamous relationships, group sex, and BDSM (Bettinger, 2014).Understanding these factors can provide valuable insights into how to address these issues in a therapeutic context.Finally, future research should also examine the reactions of therapists who work with survivors of child sexual abuse to erotic transference.As Davies and Frawley (1994) have discussed, therapists working in this area may encounter situations where there is a risk of reenactment of past abuse, as well as potential for seduction transference.Understanding the reactions of therapists in this context could be valuable for developing best practices and protocols for working with survivors of child sexual abuse, as well as informing training programs for therapists in this area and ensuring that therapists have the necessary skills and support to provide effective and ethical care to their clients.

Figure 1 .
Figure 1.eGa results of the initial and final 19-item version of the taSef.note: red -gratification; blue -fear; green -threat; orange -aversion.

Figure 4 .
Figure 4. factor-level Multi-trait Multimethod approach for convergence and discriminant validity.

Figure 5 .
Figure 5. item-level Multi-trait Multimethod approach for convergence and discriminant validity.

Table 2 .
Within-factor correlation, convergence, and discriminant scores of erotic transference clusters.

Table 3 .
Means, standard deviations, statistics, and effect sizes for the differences in erotic transference by gender, family status, sexual orientation, sex therapist, and religiosity.

Table 4 .
Correlations between taSef clusters and background measures.