Effects of Verbal and Nonverbal Communication of Affection on Avoidantly Attached Partners’ Emotions and Message Receptiveness

Abstract

Research on adult attachment in romantic relationships has focused on the negative outcomes that avoidantly attached individuals face. The present research uses observational research methods to determine if there are specific ways of communicating affection that might help avoidantly attached people reap similar levels of rewards from affectionate communication as those who are more secure. We combined three samples (N_total = 280 couples, 560 participants) who took turns describing a time they felt strong love for their partner, and coded their expressions for cues of verbal affection (i.e., emotion-laden words) and nonverbal affection (i.e., behavioral expressiveness). Higher levels of the speaker’s nonverbal affection were associated with stronger positive emotion and behavioral receptiveness (i.e., appearing engaged) for listeners higher in attachment avoidance. Altogether, we provide evidence that avoidantly attached individuals may experience positive outcomes from affectionate exchanges when the communication style is tailored to their unique needs.

Keywords

attachment romantic relationships communication emotion in relationships intimacy

Individuals who are avoidantly attached, or eschew intimacy and closeness, typically respond to intimacy with lower receptiveness and positive emotion (Hicks & Diamond, 2008) as well as more negative emotion (Mikulincer & Nachshon, 1991). Thus, an important question is how people can effectively express affection to an avoidantly attached individual. In the present research, we aim to understand if affection expressed through nonverbal and/or verbal cues can lead to positive responses among avoidantly attached individuals for whom intimacy is often difficult to achieve. We thought it possible that expressions of verbal affection might be particularly important for avoidantly attached individuals because they provide direct and overt expressions of a partner’s feelings, and suggest that their partner can be trusted in times of need. On the other hand, nonverbal affection—possibly a more covert signal of a partner’s reliability—might specifically prevent avoidantly attached individuals’ fears of intimacy from being activated.

Using observational research methods to examine partner conversations about love, we expand on the growing body of work on buffering attachment avoidance (see review by Simpson & Overall, 2014) in three main ways. First, although past work in the dyadic communication literature has mostly focused on attachment avoidance in distress situations such as conflict (Overall et al., 2013) or social support (Collins & Feeney, 2004), our work is some of the first to examine how a speaker’s affectionate messages can be most effectively communicated to avoidantly attached listeners. Second, the small body of work conducted on positive conversations (e.g., capitalization; Gosnell & Gable, 2013) has primarily focused on providing responsive support to the speaker as opposed to tailoring affectionate messages to the listener. Finally, we move beyond simply understanding the degree of intimacy-laden cues that avoidantly attached individuals desire (Stanton et al., 2017) and explore the type of affection that addresses their unique needs (i.e., enhanced evidence of the partner’s trustworthiness) to impact emotions and behavioral receptiveness during couple interactions.

Attachment Avoidance and Reduced Intimacy-Related Outcomes

Because avoidantly attached individuals often perceive close others as untrustworthy, unreliable, and uncaring (Hazan & Shaver, 1987), they often make efforts to prevent intimacy from developing and are less able to experience the positive outcomes of their partners’ affectionate behavior. At a behavioral level, one way in which avoidantly attached individuals prevent intimacy from occurring is by exhibiting less receptiveness in response to positive social stimuli relative to less avoidant individuals. We define receptiveness as a behavioral demonstration of approval, engagement, and warmth in response to social stimuli. Avoidantly attached individuals exert high levels of control over their positive emotions such as by concealing positive emotions from their partner (Feeney, 1999). Furthermore, individuals high relative to low in attachment avoidance are less responsive to their partner, even when their partner is discussing positive events (Shallcross et al., 2011). Avoidantly attached individuals’ suppression of their behavioral responses may be problematic because failure to engage with their partners’ affectionate gestures may prevent them from experiencing the full emotional benefit of these expressions.

Indeed, in addition to displaying reductions in behavioral positivity, avoidantly attached individuals also exhibit disruptions in their affective responses to intimacy-related cues, such as low levels of positive emotions and high levels of negative emotions. For example, in a gratitude induction task in which people were asked to recall a time that their relationship partner engaged in a kind behavior, those higher relative to lower on attachment avoidance were more likely to recall negative experiences involving threats and distrust, as well as less happiness and love (Mikulincer & Shaver, 2005). Even during intensely emotionally laden situations, such as spouses’ reunion after long periods of separation, avoidantly attached individuals demonstrated dampened positive emotions and higher degrees of conflict relative to those who were more securely attached (Medway et al., 1995). Taken together, avoidantly attached individuals appear to struggle to achieve positive behavioral and affective outcomes in response to intimacy-related cues.

Partner Buffering of Attachment Avoidance

Because avoidantly attached individuals have unique needs (i.e., they require enhanced evidence of the partner’s trustworthiness), perhaps affection is best communicated in tailored ways in order for them to display heightened behavioral receptiveness and experience more positive and less negative emotions. Recent findings on partner buffering, or reducing the negative outcomes associated with attachment avoidance, suggest that relationship strategies that address an individual’s unique needs are more effective at helping romantic partners achieve positive relationship outcomes (Simpson & Overall, 2014).

While much of the initial buffering research has focused on reducing negative outcomes in times of conflict (e.g., Farrell et al., 2016; Overall et al., 2013), recent research suggests that it is also important to communicate to avoidantly attached individuals in tailored ways during positive interactions. In support conversations, as partners increased their levels of support from low to moderate levels, avoidantly attached recipients showed poorer outcomes (increases in distress and perceived control/criticism), but this trend reversed as support continued to increase to high levels (Girme et al., 2015). Thus, in the face of unwavering support, avoidantly attached individuals become better convinced of their partners’ reliability, allowing them to lower their guard sufficiently to reap the benefits of the supportive behavior. Furthermore, perceptions that the partner is engaging in positive relationship behaviors, such as expressing kind words or compliments (Stanton et al., 2017) or expressing gratitude (Park et al., 2019) increases relationship quality and positive emotions and reduces negative emotions among avoidantly attached individuals. These findings suggest that communication patterns that address avoidantly attached individuals’ lack of trust, through clear unwavering support or providing a signal that the avoidantly attached partner is cared for, may improve personal and relationship outcomes.

Communicating Verbal and Nonverbal Affection to Avoidantly Attached Individuals

Although intimacy-laden cues tend to promote personal and relationship well-being for avoidantly attached individuals (Stanton et al., 2017), it remains unclear which types of cues are likely to have these effects. Given that affection can be expressed through both verbal and nonverbal channels (Andersen et al., 2006), we examined the influence of verbal and nonverbal expressions of affection on the reactions of romantic partners to such expressions. To capture the wide variety of verbal and nonverbal affectionate behaviors demonstrated by couples in the present research, we developed our own coding scheme. We took an iterative approach to developing the codes, in which we first drew upon the literature on observational coding of couples in affectionate contexts (e.g., Roberts & Greenberg, 2002; Tucker & Anders, 1998) and selected the codes that we saw present in our sample. We then developed our own codes for the remainder of the themes that we observed in the sample that were not developed in previous literature. Although we did not have a priori predictions as to whether one or both of the channels (verbal and nonverbal) would be particularly beneficial, below we highlight the theoretical rationale for why each channel may or may not buffer the negative personal and relational outcomes of avoidantly attached individuals in affectionate situations.

Verbal statements of affection include words and phrases such as “I love you” or “I like you” (Floyd, 2018). People often communicate affection through verbal channels when they wish to be overt and unambiguous about the state of their relationship (i.e., that their relationship is characterized by love; Floyd, 1997). Because avoidantly attached individuals tend to respond to clear, unwavering cues that their partner cares about them (Girme et al., 2015), verbal statements may help avoidantly attached individuals to accept the veracity of their partners’ affections. However, it is also possible that communicating affection through verbal channels may backfire for avoidantly attached listeners. Verbal channels relative to nonverbal channels are more likely to include deceptive communication in pursuit of goals such as to pressure a recipient into heightened levels of relationship commitment (Booth-Butterfield & Trotta, 1994). Because avoidantly attached individuals resist intimacy, especially when it is not autonomously chosen (Mikulincer & Shaver, 2007), any doubts about the sincerity of their partner’s affections may prevent avoidantly attached individuals from realizing the full personal and relational benefits of the expression.

We refer to nonverbal signals of intimacy as nonverbal affection, which represent the degree to which a person is an engaged and active participant in a social interaction (Coker & Burgoon, 1987). Nonverbal affection cues include a wide range of behaviors including smiling, forward leans, head nods, and animated voice and gestures (Andersen et al., 2006). Nonverbal affection may be particularly beneficial for avoidantly attached individuals because its perceived genuineness and relatively indirect style of communication may ameliorate their difficulties with trust and tendency to eschew closeness. Nonverbal affection cues are thought to be particularly genuine signals of affection because they arise spontaneously and are consistent across multiple channels (e.g., vocal animation, facial expressions), making the speaker’s affectionate signal more reliable (Andersen et al., 2006). Thus, nonverbal affection cues may target avoidantly attached individuals’ difficulties with trust (Mikulincer & Shaver, 2007) by providing converging evidence of their partners’ affections. At the same time, nonverbal affection cues (compared to verbal ones), while powerful, are relatively indirect forms of emotional expression, and there are a wide variety of interpretations for these behaviors. For example, some cues of nonverbal affection (e.g., being vocally and facially animated) could express love, liking, or simply joy. Thus, this relatively indirect form of communication may prevent avoidantly attached individuals’ fears of closeness from being activated, while still providing the benefits of affectionate communication. On the other hand, there may be reasons why nonverbal affection cues may not buffer the poor personal and relational outcomes of avoidantly attached individuals. Because nonverbal affection cues are relatively indirect, avoidantly attached individuals may not correctly infer that their partner’s behavior is indeed affectionate. Avoidantly attached individuals typically ignore signs of closeness and intimacy (Mikulincer & Shaver, 2007), and thus nonverbal affection signals may not be communicating affection clearly enough for these individuals to reap the benefits.

Overview of the Present Research

The primary aim of the current investigation is to understand how affection can be most effectively communicated to avoidantly attached individuals so that they receive their partners’ messages of love. We used observational research methods in the context of affectionate communication to examine which speaker behaviors were most strongly associated with positive outcomes for avoidantly attached listeners across two North American countries (United States and Canada). After assessing attachment style, we video recorded couples discussing a time the speaker felt strong love for the listener and measured self-reports of both couple members’ positive and negative emotions after the conversation. Trained coders made ratings on the speaker’s verbal and nonverbal cues of affection for the listener, as well as how well the listener appeared to receive these cues. We then used these codes to predict how verbal and nonverbal cues of affection mapped onto avoidantly attached listeners’ behavioral receptiveness to these messages, as well as their self-reported positive and negative emotions.

Methods

Participants

Sample 1

We recruited 100 couples (200 participants) from undergraduate psychology courses and community sources (Kijiji.com, Craigslist.org, Facebook groups, and campus flyers) in a large Canadian city. Participants either received one course credit or $20 CAD. Of the 100 couples, 90 were heterosexual, 8 were lesbian, and 2 were gay male couples. The age of participants ranged from 18 to 47 years (M = 21.97, SD = 4.99). Couples were dating for a minimum of less than 1 month and a maximum of 7 years (M = 17.92 months, SD = 16.96 months). Roughly a quarter (25.5%) of couples reported living together.

Sample 2

We recruited 124 couples (248 participants) from the same institution and the surrounding community as Study 1 using identical advertising procedures. Participants received $20 to complete the study. Of the 124 couples, 118 were heterosexual and 6 were lesbian. Ages ranged from 18 to 46 years (M = 21.70 years, SD = 4.19 years). Couples were together for a minimum of 2 months and a maximum of 10 years (M = 24.66 months, SD = 22.07 months). Almost a third (29.8%) of the couples lived together at the time of study.

Sample 3

We recruited 80 couples (160 participants) from the San Francisco Bay Area by flyers posted throughout the area and online via Craigslist.org. Of the 80 couples, 75 were heterosexual, 4 were lesbian, and 1 was a gay male couple. The couples had been dating for a minimum of 6 months and a maximum of 30 years (M = 29.21 months; SD = 43.40 months). Participants ranged in age from 18 to 60 years (M = 23.84, SD = 6.37). In addition, 48% of the couples were cohabitating. We compensated all participants $20 USD for participating in the study. The data from Sample 3 were part of a larger study on couple communication (Impett et al., 2010).

Procedure

In all samples, we invited couples to come into the lab and first asked them to complete background questionnaires followed by videotaped conversations. All participants provided informed consent to be videotaped. Specifically, we instructed couples to describe a time they “felt a lot of love for their partner and how they expressed it”. The listener was not given specific instructions, and thus was able to react freely. Each partner had the opportunity to be in the role of the “speaker” and the role of the “listener.” The order of speech was determined by a random number generator (Studies 1 and 2) or a coin toss (Study 3). The mean length of the conversation was 2 min, 7 s (SD = 1 min, 46 s), with a range of 15 s to 17 min, 30 s. Following each conversation, both members of the couple provided ratings of their positive and negative emotions.

Measures

Background measures

Participants provided responses to basic demographic information (i.e., gender, age, relationship duration).

Attachment

In Samples 1 and 2, adult attachment was measured using the Experiences in Close-Relationships–Revised questionnaire (ECR-R; Fraley et al., 2000). Participants answered questions assessing attachment anxiety (e.g., “I’m afraid that I will lose my partner’s love,” 18 items, α =.91, M = 2.85, SD = 0.98) and avoidance (e.g., “I prefer not to be too close to romantic partners,” α = .91, 18 items, M = 2.31, SD = 0.83), on a 7-point scale (1 = strongly disagree to 7 = strongly agree), with higher scores reflecting greater attachment anxiety and avoidance. In Sample 3, participants completed the Experiences in Close Relationships scale (ECR; Brennan et al., 1998), with questions about attachment anxiety (e.g., “I worry that others won’t care about me as much as I care about them,” 18 items, α = 0.89, M = 2.81, SD = 0.58), and avoidance (e.g., “I am nervous when another person gets too close to me,” α = 0.90, 18 items, M = 2.03, SD = 0.57) on a 5-point scale (1 = strongly disagree to 5 = strongly agree). Scores on the ECR and ECR-R were standardized before combining to ensure values were comparable for analysis.

Relationship satisfaction

Relationship satisfaction was assessed with five items such as “Our relationship makes me happy” (Rusbult et al., 1998). In Samples 1 and 2, participants responded to these items on a 9-point scale (1 = strongly disagree to 9 = strongly agree; α = .86, M = 7.63, SD = 1.17), whereas in Sample 3, they made ratings on a 7-point scale (1 = strongly disagree to 7 = strongly agree; α = .88, M = 6.13, SD = 0.93). Scores on this measure across samples were standardized before combining to ensure the values were comparable for analysis.

Self-reported emotions

Participants rated their emotions at the beginning of the lab session before the conversations (baseline), and immediately after each love conversation. Based on prior research (Impett et al., 2010) participants rated the extent to which they experienced six positive emotion triplets (happy/pleased/joyful, affectionate/loving/caring, proud/good about self, compassionate/sympathetic, grateful/appreciative, and cared about/loved/connected) and five negative emotions (anxious/nervous, lonely/isolated, angry/irritated/hostile, contempt/disgust with partner, and disappointed/let down) on a 7-point scale (1 = not at all to 7 = a lot). The scales were reliable for both positive (α =.84, M = 5.33, SD = 0.97) and negative emotions (α =.82, M = 1.65, SD = 0.87) before the conversation (baseline), as well as positive (α = .92, M = 6.11, SD = 0.90) and negative emotions (α = .79, M = 1.18, SD = 0.47) after the conversation.

Coding conversation behaviors

A team of four to six undergraduate coders who were blind to the study’s research questions independently provided ratings of the videotaped speaker and listener behaviors. We included codes from previous research on couples’ conversations that included elements of verbal or nonverbal communication of affection or resistance to affectionate expression (Roberts & Greenberg, 2002; Tucker & Anders, 1998). We added additional codes (e.g., authenticity and engagement) that we felt had specific attachment relevance and emerged from watching videos of the couples’ interactions during the lab task (see Table 1 for code sources). We adopted Shapiro and Gottman’s (2004) approach to coding, which is a blend of the physical features and cultural informants approach. The physical features approach provides descriptions and exemplars of features to look for that which are associated with that code. The cultural informants approach allows the coder to bring in their subjective experience to decode the participant’s behavior in the broader context of the conversation and couple dynamic. This integrated approach allows researchers to design codes that combine objective and subjective cues to attain reliability while maintaining greater ecological validity.

Table 1.

Descriptions and Reliability of Speaker and Listener Codes.

Code	Description of code	Modeled after	Intraclass correlation
Nonverbal affection (speaker)
In love	How “in love” the speaker looked with the listener, including warm smiles, baby voices, glistening eyes, head cocked to the side.	“Surge Love” (Roberts & Greenberg, 2002)	.79
Engagement	How engaged the speaker looked with the listener, including maintenance of eye contact, body orientation pointed toward the partner, touch or trying to involve the partner in the conversation.	Code designed for the present research	.83
Enjoyment	How much the speaker looked as if they were enjoying the conversation, including use of humor and laughter.	“Enjoyment” (Tucker & Anders, 1998)	.79
Expressiveness	How behaviorally expressive the speaker appeared, including shifts in body language, use of gestures, facial expressions, and fluctuations in vocal pitch.	“Expressiveness” (Tucker & Anders, 1998)	.79
Tension^a	How tense the speaker appeared in the interaction, including fidgeting, awkward/strained voice or facial features, nervous laughter, difficulty knowing what to say, and frequently shifting posture or gaze.	“Tenseness” (Tucker & Anders, 1998)	.72
Verbal affection (speaker)
Emotional content	How freely the speaker discussed their emotions surrounding their disclosure, above and beyond factual descriptions of the event.	“Vulnerable Disclosure” (Roberts & Greenberg, 2002)	.79
Authenticity	How authentic the speaker appeared to be when communicating their emotions for the partner. Expressions are sincere and earnest (rather than unserious or sarcastic).	Code designed for the present research	.71
Deflecting	How much the speaker deflected the conversation away from the moment they felt love. The speaker may stray from the task and discuss other topics.	“General Sharing” (Roberts & Greenberg, 2002)	.85
Security^a	How much the speaker focused on the partner’s impact on the self during the conversation. The speaker may discuss how the partner makes them feel, or how they are cared for or supported by the partner.	“Feelings of security, comfort and trust” (Roberts & Greenberg, 2002)	.72
Cold^a	How cold the speaker appeared to be in the conversation. The extent to which the speaker is defensive, closed off, uses a harsh tone of voice or communicates lack of respect for the partner. This can take the form of insults, blame, sarcasm, or hostile humor.	“Contempt” (Roberts & Greenberg, 2002)	.83
Receptiveness (listener)
Love	How loved the listener looked in the interaction. Includes indicators such as warm smiles, baby voices, glistening eyes, head cocked to the side.	“Surge Love” (Roberts & Greenberg, 2002)	.86
Withdrawal	How withdrawn the listener appeared with the speaker. The listener does not appear to accept the speaker’s message, may turn their head or body away from the speaker and avoid eye contact.	“Withdrawal” (Overall, Simpson, & Struthers, 2013)	.90
Cold	How cold the listener appeared to be in the conversation. The listener does not convey sensitivity to the message the speaker was trying to convey, responds in a short, harsh or sarcastic way, or uses insults or hostile humor.	“Contempt” (Roberts & Greenberg, 2002)	.70

Note. ^aItem excluded due to low communality on factor analysis.

Although we were primarily interested in the global assessment of the speaker’s verbal and nonverbal communication of affection, we designed specific verbal and nonverbal speaker codes to allow for the possibility that individual types of behaviors may not necessarily load together, and thus have a unique effect on the listeners’ responses. We subsequently performed an exploratory factor analysis (EFA; detailed below). The verbal affection speaker codes designed to capture the verbal depth of the message included (a) communicating emotional content (intraclass correlation [ICC] = .79), (b) authenticity of words (ICC = .71), (c) deflecting communication of affectionate words (reverse coded, ICC = .85), and (d) communicating feelings of security (ICC = .72). The nonverbal affection speaker codes designed to capture the extent to which an individual is an engaged and active participant in a social interaction included (a) engagement (ICC = .83), (b) enjoyment (ICC = .79), (c) expressiveness (ICC = .79), (d) tension (reverse coded, ICC = .79), (e) coldness (reverse coded, ICC = .83), and (f) looking “in love” (ICC = .79). The listener receptiveness codes were intended to capture the listeners’ behavioral acceptance of the message included (a) withdrawal (ICC = .90), (b) coldness (ICC = .70), and (c) looking loved (ICC = .86). See Table 1 for full description of speaker and listener codes presented to the coders to make their ratings.

We assembled a total of three undergraduate coder groups and assigned each group to make ratings for one of the three samples. We provided all coders with a detailed coding scheme and coders attended four to six 2-hr intensive training sessions to ensure interrater reliability. Coders rated all questions with the following metric: 1 to 2 = low, 3 to 5 = moderate, 6 to 7 = high, and were asked to keep in mind the frequency, intensity, and duration of each behavior (Overall et al., 2013). In the initial session, the first author provided in-depth instructions to the coders to describe each of the behaviors present in each category. After the verbal descriptions, the first author provided video examples to the coders to demonstrate high, medium, and low levels of each of the categories. The session concluded with the coders making independent ratings of six videos (three speakers and three listeners) to practice the coding scheme. After each video, all members revealed their ratings and were asked to discuss their rationale for making their selections. The first author confirmed correctly identified behavior and re-directed incorrect responses. We gave coders practice homework assignments and the coding team met for three additional follow-up sessions to ensure reliability of each category. After reliability in the training session was achieved at an intraclass correlation of .70 or higher (the benchmark for “very good” reliability; Mitchell, 1979), coders individually completed ratings for all videos within the sample, which were averaged together.¹ Ratings were made for every 30-s conversation segment for both the speaker and the listener categories.

Exploratory Factor Analysis

Once all coding was complete, we combined the data from all three samples to increase statistical power. We then conducted an Exploratory Factor Analysis (EFA) on the speaker and listener behaviors to assess how the codes loaded together (see Table 2 for factor analysis). We opted to perform one factor analysis on both the speaker and listener codes, because there were insufficient listener codes to perform a separate factor analysis. We performed the EFA with the maximum likelihood extraction method and promax rotation to achieve simple structure and assume correlations among the factors. We retained items with a factor loading of .50 or higher (Comrey & Lee, 1992; Costello & Osborne, 2005) and dropped items with cross loadings above .40.

Table 2.

Factor Loadings for Speaker and Listener Codes (Combined Sample).

Codes	Nonverbal affection (speaker)	Verbal affection (speaker)	Receptiveness (listener)
Emotion (speaker)	−.05	.84	.04
Expressiveness (speaker)	.70	.16	.15
Enjoyment (speaker)	.82	−.11	−.02
In love (speaker)	.67	.21	−.19
Authenticity (speaker)	.29	.70	.06
Engagement (speaker)	.81	−.14	−.02
Tension (speaker)	−.38	.07	.02
Deflection (speaker, reverse coded)	.38	−.70	.16
Security (speaker)	.06	.45	.09
Cold (speaker, reverse coded)	−.07	−.18	.33
Love (listener)	.21	−.05	−.66
Withdrawal (listener, reverse coded)	.08	.07	.72
Cold (listener, reverse coded)	.05	.01	.64

Note. Bolded items were retained for analysis.

Results of the EFA yielded a three-factor solution of speaker and listener behaviors, which explained 66.38% of the variance. We labeled the first factor speaker nonverbal affection, which included the codes “engagement,” “enjoyment,” “expressiveness,” and “in love.” We labeled the second factor speaker verbal affection, which comprised the codes “emotion,” “authenticity,” and “deflection” (reverse coded). We labeled the third factor listener receptiveness, which is comprised of “love,” “withdrawal” (reverse coded), and “cold” (reverse coded). We dropped the speaker tension, cold, and security codes due to failure to meet the minimum loading requirement.

Results

Effects of Fluctuations in Speaker Verbal Affection and Nonverbal Affection for Avoidantly Attached Listeners’ Behavioral Receptiveness: Within-Person Analyses

Our first set of analyses examined whether speaking partners were able to attenuate avoidant listeners’ low levels of behavioral receptiveness through the use of affectionate communication. We tested this idea by analyzing the repeated measures of the speaker’s verbal and nonverbal affection and the listeners’ receptiveness at each 30-s interval to examine if increases in the speaker’s affectionate communication were associated with increases in the listeners’ displays of receptiveness across the conversation. These analyses examine how fluctuations within the listeners’ receptiveness correspond with fluctuations in their speaking partner’s verbal (M within-person SD = 0.31, range of within-person SD = 0–1.04) and nonverbal affection (M within-person SD = 0.15, range of within-person SD = 0–0.76).

We analyzed the data with multilevel modeling (linear mixed model function) in SPSS 26. We tested a two-level cross model with random intercepts in which persons are nested within dyads, and person and conversation segments are crossed to account for the fact that both partners were in the conversation at the same time (Kenny et al., 2006). To capture the unique within-person variance, we person-centered our level-1 speaker communication styles (e.g., within-person fluctuations in speaker verbal and nonverbal affection) and controlled for between-person variance with the level-2 aggregated communication patterns (e.g., speaker verbal affection and nonverbal affection over the course of each speaker’s entire conversation), which we grand mean centered. We also grand mean centered the attachment variables. We were primarily interested in testing whether attachment avoidance was associated with lower levels of behavioral receptiveness, and whether within-person fluctuations in speaker communication patterns attenuated these effects. As such, we examined the main effects of attachment avoidance, as well as the cross-level interactions between attachment avoidance and speaker verbal affection and between attachment avoidance and speaker nonverbal affection predicting listener receptiveness. To control for the effects of attachment insecurities more broadly, we also entered the main effects and cross-level interactions between attachment anxiety and speaker verbal affection, and between speaker anxiety and nonverbal affection simultaneously into the model. To prevent the within- and between-person variance from becoming conflated, we also entered the same-level interactions between the aggregated speaker communication styles (i.e., verbal affection and nonverbal affection) and listener attachment styles (i.e., avoidance and anxiety) simultaneously into the model. Finally, because we joined multiple datasets for the present research, we followed the recommendations of Curran and Hussong (2009) and conducted fixed-effects Integrative Data Analysis (IDA), in which sample was effects coded and entered as a fixed effect. To control for sample, we created two different effects codes. For the first effects code, the three different samples were coded as follows: Sample 1 = 1, Sample 2 = 0, and Sample 3 = -1. For the second effects code, the three samples were coded as follows: Sample 1 = 0, Sample 2 = 1, and Sample 3 = -1.

Verbal affection

As shown in Table 3, higher scores on attachment avoidance were associated with less listener receptiveness. There was a negative effect of speaker verbal affection on listener receptiveness, suggesting that the more a speaker elaborates on their affectionate feelings, the less receptive the listener was in the conversation. There was no speaker verbal affection by listener avoidance interaction, indicating that speaker verbal affection had similar effects in eliciting behavioral receptiveness in highly avoidant relative to less avoidant listeners.

Table 3.

Speaker Nonverbal Affection and Verbal Affection Predicting Listener Receptiveness (Combined Sample).

	Within-person analyses					Between-person analyses
Measure	b	SE	t	r	95% CI	b	SE	t	r	95% CI
Listener avoidance	−0.04	0.01	−3.37***	0.08	[−0.07, −0.02]	−0.05	0.02	−2.31*	0.10	[−0.09, −0.01]
Listener anxiety	−0.02	0.01	−1.98*	0.05	[−0.05, −0.01]	−0.02	0.02	−1.10	0.05	[−0.06, 0.02]
Speaker nonverbal affection	0.32	0.03	10.84***	0.24	[0.26, 0.38]	0.37	0.04	8.27***	0.39	[0.28, 0.46]
Speaker verbal affection	−0.05	0.02	−3.45***	0.09	[−0.08, −0.02]	0.02	0.04	0.39	0.02	[−0.07, 0.10]
Sample Effects Code 1	0.17	0.05	3.72***	0.25	[0.08, 0.26]	−0.07	0.03	−2.46*	0.18	[−0.13, −0.01]
Sample Effects Code 2	−0.06	0.04	−1.46	0.10	[−0.15, 0.02]	0.05	0.03	1.77^†	0.13	[−0.01, 0.10]
Speaker nonverbal affection x listener avoidance	0.06	0.03	2.14*	0.05	[0.01, 0.12]	0.11	0.05	2.42*	0.11	[0.02, 0.20]
Speaker nonverbal affection x listener anxiety	0.01	0.03	0.23	0.005	[−0.06, 0.07]	−0.02	0.05	−0.32	0.01	[−0.11, 0.08]
Speaker verbal affection x listener avoidance	−0.02	0.02	−1.30	0.03	[−0.06, 0.01]	0.01	0.04	0.15	0.01	[−0.07, 0.09]
Speaker verbal affection x listener anxiety	0.02	0.02	0.93	0.02	[−0.02, 0.05]	0.01	0.04	0.33	0.02	[−0.07, 0.10]

Note. Effect sizes were computed using Rosenthal and Rosnow’s (2007) formula: r = √(t²/t² + df). CI = confidence interval.

†

p < .10. *p < .05. **p < .01. *** p < .001.

Nonverbal affection

In moments when speakers were rated by outside observers as being higher (relative to lower) in nonverbal affection, listeners were seen as more receptive in the conversation during those same moments. However, most critically and as shown in Figure 1A, the speaker’s nonverbal affection by listener avoidance interaction was significant, suggesting that the effect of avoidance on listener receptiveness in any particular moment depended on the degree of nonverbal affection their speaking partner appeared to enact at that time. Simple slopes analyses revealed that in moments when speakers were low in nonverbal affection (1 SD below their own mean), listeners high in attachment avoidance were coded as less receptive by outside observers relative to those low in attachment avoidance, b = -.06, SE = .01 t(1,972) = -4.01, p < .001. However, in moments when speakers displayed high levels of nonverbal affection (1 SD above their own mean), the negative effect of listener avoidance on receptiveness was weakened, such that the slope dropped to marginal significance, b = -.03, SE = .01, t(2,021) = -1.84, p = .07. Analyzing the simple slopes the other way, in moments when speakers were higher (relative to lower) in nonverbal affection, listeners were significantly more likely to be receptive at that time regardless of whether they were low, b = .26, SE = .04, t(2,031) = 6.18, p < .001, or high in attachment avoidance, b = .38, SE = .04, t(2,040) = 9.12, p < .001.

Figure 1.

Speaker nonverbal affection by listener avoidance interactions predicting within (A)- and between (B)-person listener receptiveness, and post-conversation positive emotions (C) for the combined sample.

Effects of Speaker Verbal Affection and Nonverbal Affection for Avoidantly Attached Listeners’ Behavioral and Emotional Reactions: Between-Person Analyses

Next, we tested our buffering predictions at the between-person level by aggregating speaker verbal affection and nonverbal affection, as well as listener receptiveness, across the entire conversation. Thus, the benchmark for testing the effectiveness of the speakers’ high levels of verbal affection and nonverbal affection is the amount of receptiveness exhibited by listeners high relative to low in attachment avoidance. We also used the speakers’ aggregate verbal affection and nonverbal affection to predict (avoidant) listeners’ positive and negative emotions at the end of the conversation, which was assessed at a single timepoint post-discussion and could be analyzed only at the between-person level.

Using the linear mixed model function in SPSS 26, we conducted three separate two-level multilevel models (to account for the interdependence between couple members) in which we predicted the (a) listeners’ receptiveness, (b) positive emotions, and (c) negative emotions from the listeners’ avoidance, the speaker’s average verbal affection and nonverbal affection across the conversation and the interactions between the speakers’ verbal affection and listener avoidance as well as the speakers’ nonverbal affection and the listeners’ attachment avoidance. We also included the main effects and interactions between the listeners’ attachment anxiety and the aggregates of speaker verbal affection and nonverbal affection to control for the effects of attachment insecurity more broadly. We grand-mean centered all predictors.

Verbal affection and avoidant listeners’ behavioral receptiveness

As already demonstrated and shown in Table 3, higher listener attachment avoidance was associated with less behavioral receptiveness. Speaker verbal affection was unrelated to the listeners’ level of behavioral receptiveness, regardless of the listeners’ level of attachment avoidance. These results indicate that speaker verbal affection has a similar effect on the listening partner’s receptiveness in the conversation for those low and high in attachment avoidance.

Verbal affection and avoidant listeners’ positive emotions

As shown in Table 4, the higher a participant scored on attachment avoidance, the lower their ratings on post-conversation positive emotions. There was no main effect of speaker verbal affection on listener positive emotions, and no evidence of a speaker verbal affection by listener avoidance interaction. Thus, speaker verbal affection was not associated with heightened positive emotions for those low or high in attachment avoidance.

Table 4.

Speaker Nonverbal Affection and Verbal Affection Predicting Listener Positive Emotions (Combined Sample).

	Between-person analyses
Measure	b	SE	t	r	95% CI
Listener avoidance	−0.16	0.04	−3.76***	0.17	[−0.25, −0.08]
Listener anxiety	−0.15	0.04	−3.45***	0.15	[−0.23, −0.06]
Speaker nonverbal affection	0.24	0.10	2.46*	0.11	[0.05, 0.42]
Speaker verbal affection	0.08	0.09	0.88	0.04	[−0.10, 0.26]
Sample Effects Code 1	0.18	0.07	2.59*	0.15	[0.04, 0.31]
Sample Effects Code 2	0.22	0.06	3.49***	0.20	[0.10, 0.35]
Speaker nonverbal affection x listener avoidance	0.17	0.09	1.81^†	0.08	[−0.01, 0.36]
Speaker nonverbal affection x listener anxiety	−0.12	0.10	−1.20	0.05	[−0.31, 0.08]
Speaker verbal affection x listener avoidance	−0.05	0.08	−0.57	0.02	[−0.21, 0.12]
Speaker verbal affection x listener anxiety	0.18	0.09	1.97*	0.09	[0.01, 0.36]

Note. Effect sizes were computed using Rosenthal and Rosnow’s (2007) formula: r = √(t²/t² + df). CI = confidence interval.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Verbal affection and avoidant listeners’ negative emotions

As shown in Table 5, avoidantly attached individuals reported significantly more negative emotions after their partners’ expression of affection relative to those low in attachment avoidance. The more verbal affection the speaker expressed, the less negative emotions the listener reported. Nevertheless, there was no speaker verbal affection by listener avoidance interaction. This finding suggests that although verbal affection did not fully buffer the effect of attachment avoidance on negative emotions, verbal affection was still associated with lower negative emotions for all participants on average, regardless of attachment avoidance.

Table 5.

Speaker Nonverbal Affection and Verbal Affection Predicting Listener Negative Emotions (Combined Sample).

	Between-person analyses
Measure	b	SE	t	r	95% CI
Listener avoidance	0.06	0.02	2.57*	0.11	[0.01, 0.11]
Listener anxiety	0.13	0.02	5.64***	0.24	[0.09, 0.18]
Speaker nonverbal affection	−0.02	0.05	−0.33	0.02	[−0.11, 0.08]
Speaker verbal affection	−0.14	0.05	−2.89**	0.14	[−0.23, −0.04]
Sample Effects Code 1	0.02	0.03	0.69	0.04	[−0.04, 0.09]
Sample Effects Code 2	0.04	0.03	1.38	0.08	[−0.02, 0.10]
Speaker nonverbal affection x listener avoidance	0.02	0.05	0.39	0.02	[−0.08, 0.12]
Speaker nonverbal affection x listener anxiety	−0.04	0.05	−0.68	0.03	[−0.14, 0.07]
Speaker verbal affection x listener avoidance	−0.08	0.05	−1.67	0.07	[−0.17, 0.01]
Speaker verbal affection x listener anxiety	−0.07	0.05	−1.45	0.06	[−0.17, 0.03]

Note: Effect sizes were computed using Rosenthal and Rosnow’s (2007) formula: r = √(t²/t² + df). CI = confidence interval.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Nonverbal affection and avoidant listeners’ behavioral receptiveness

Furthermore, when speakers demonstrated more (vs. less) nonverbal affection, listeners were more receptive during the conversation. However, and as shown in Figure 1B, a significant speaker nonverbal affection by listener avoidance interaction revealed that this effect was moderated by speaker nonverbal affection.

Simple slopes analyses indicated that when speakers were low in nonverbal affection, listeners high in attachment avoidance were coded as less receptive by outside observers, b = -.10, SE = .03 t(486) = -3.58, p < .001, but when speakers displayed high levels of nonverbal affection, listeners high in attachment avoidance were seen as just as receptive as less avoidantly attached individuals, b = .01, SE = .03, t(483) = 0.16, p = .88. In addition, when speakers demonstrated high (relative to low) levels of nonverbal affection during the conversation, listeners were significantly more receptive regardless of whether they were low, b = .26, SE = .06, t(470) = 4.13, p < .001, or high in attachment avoidance, b = .48, SE = .06, t(449) = 7.37, p < .001.

Nonverbal affection and avoidant listeners’ positive emotions

Speakers who were high levels of nonverbal affection had listening partners who reported more positive emotions relative to those speakers who were low in nonverbal affection. In addition, and as shown in Figure 1C, there was a significant speaker nonverbal affection by listener avoidance interaction, indicating that avoidantly attached listeners’ positive emotions depended on the level of nonverbal affection their speaking partner demonstrated during the conversation.

Simple slopes analyses revealed that when speakers were coded as low in nonverbal affection, highly avoidant listeners reported significantly less positive emotions than less avoidant listeners, b = -.24, SE = .06, t(492) = -4.24, p < .001. However, highly avoidant listeners reported similarly high levels of positive emotions as less avoidant listeners when the speaker high levels of nonverbal affection, b = -.08, SE = .07, t(500) = -1.21, p = .23. Analyzing the simple slopes another way, highly avoidant listeners reported significantly more positive emotions when speakers were rated as high relative to low in nonverbal affection, b = .41, SE = .14, t(513) = 2.97, p = .003. However, listeners low in attachment avoidance reported similar levels of positive emotions when speakers were rated as high versus low in nonverbal affection, b = .07, SE = .13, t(513) = 0.49, p = .62.

Nonverbal affection and avoidant listeners’ negative emotions

There was no main effect of speaker nonverbal affection on negative emotions, nor a speaker nonverbal affection by listener avoidance interaction, suggesting that levels of speaker nonverbal affection were equally ineffective at keeping listener negative emotions low, regardless of listeners’ levels of attachment avoidance.

Ruling Out Alternative Explanations and Providing Evidence for Generalizability

We next sought to rule out potential alternative explanations for the buffering effect of speaker nonverbal affection for avoidantly attached listeners. First, we sought to ensure that the effects of speaker nonverbal affection for avoidantly attached individuals were not due to pre-existing differences in relationship satisfaction. The results held at conventional or marginal significance when controlling for the baseline relationship satisfaction of the speaker (.18≥ all bs ≥ .06, .02 ≤ all ps ≤ .06) and the listener (.24≥ all bs ≥ .08, .03 ≤ all ps ≤ .08). Second, it was also possible that our effects could be attributed to participants experiencing a high degree of positive emotions prior to arriving in the lab, rather than to the behavioral patterns during the conversation. The results also held when controlling for the listeners’ baseline positive emotions with one exception. The speaker nonverbal affection by listener avoidance interaction dropped below (marginal) significance for listener positive emotion when controlling for the listeners’ baseline positive emotion, b = .12, SE = .09, t(491) = 1.37, p = .17, though the pattern remained the same as demonstrated in Figure 1C. Third, we reasoned that some individuals may demonstrate heightened receptiveness and positive emotions not due to any speaking style in particular, but because they spent more time on the intimacy-inducing task. The results remained the same at conventional or marginal significance when controlling for the length of the conversation (.16 ≥ all bs ≥ .08, .03 ≤ all ps ≤ .10). Finally, we aimed to rule out the possibility that the speaker nonverbal affection by listener avoidance interaction may be observed due to differing communication patterns by couples at varying relationship length. Controlling for relationship length did not change the observed effects (.17 ≥ all bs ≥ .06, .02 ≤ all ps ≤ .08).

We then examined several moderators to test the generalizability of our effects. In separate models, we tested if the effects of speaker nonverbal affection for avoidantly attached listeners were moderated by verbal affection, gender, relationship length, age, the order in which participants spoke or listened, and the sample from which participants were drawn. Out of the 21 possible three-way interactions, 4 were marginally or conventionally significant. However, none of the interactions appeared across all three outcome variables for which we obtained the nonverbal affection by avoidance interaction, suggesting these variables do not reliably alter the aforementioned results. See Supplemental Materials for the full set of moderation analyses.

Discussion

In the current investigation, we examined the specific types of affectionate communication behaviors that evoke positive outcomes in avoidantly attached listeners. Through coding and factor analyzing the conversations, we identified two distinct clusters of behaviors through which speakers can communicate affection to their partner: verbal and nonverbal affection. Nonverbal affection was associated with a number of positive outcomes in general. Listeners high in attachment avoidance showed particular benefits when speakers demonstrated greater nonverbal affection and were more negatively affected by low use of nonverbal affection relative to less avoidantly attached listeners. At high levels of speaker nonverbal affection, avoidantly attached listeners reported positive emotion and listener receptiveness that were equally high as listeners low in attachment avoidance. At low levels of speaker nonverbal affection, highly avoidant listeners reported significantly less positive emotion and receptiveness relative to those who were low in attachment avoidance. Thus, in comparison to those lower in attachment avoidance, highly avoidant listeners were more sensitive to the relative degree of nonverbal affection cues.

Although avoidantly attached individuals generally reported positive outcomes in response to speaker nonverbal affection cues, one exception was in their reports of negative emotion. Speaker nonverbal affection was not associated with low listener negative emotion for all participants on average, and this pattern did not differ for individuals high in attachment avoidance. As such, speaker nonverbal affection was not associated with negative emotions for individuals either high or low in attachment avoidance (although levels of negative emotions were relatively low for all participants in our samples). Instead, speaker verbal affection was associated with low levels of negative emotions. Although high levels of speaker verbal affection were associated with low negative emotions on average, verbal affection did not fully buffer avoidantly attached individuals’ negative emotions. However, we did not see evidence of verbal affection being (positively) associated with other desirable outcomes.

Theoretical Mechanisms for Avoidant Responses to Nonverbal Affection Cues

The reasons why avoidantly attached individuals responded most strongly to different affectionate cues may well be related to their negative working models of others. Speaker nonverbal affection may have been particularly beneficial for avoidant listeners because avoidantly attached individuals’ interpersonal skepticism may require reliable signals of love to challenge these negative working models of others. That is, because individuals high in attachment avoidance believe that others are untrustworthy (Pistole, 1994), they may need sufficiently strong cues that others are reliable in order for them to let their guard down. Indeed, behavioral expressions are typically interpreted as the most trustworthy form of communication, particularly when there is an incongruence between behavioral and verbal messages (Gupta, 2013). As such, nonverbal affection may evoke positive reactions in avoidantly attached individuals because it is most likely to be a genuine signal of affection. Thus, nonverbal affection cues may be sufficiently strong to overcome avoidantly attached individuals’ barriers and challenge their negative working models of others.

Another potential explanation for the benefits of communicating love to avoidantly attached individuals with high levels of nonverbal affection is that such behaviors are a comparatively more indirect form of communication than verbally sharing one’s feelings. Such off-record communication may be less likely to activate the attachment system than direct verbal expressions. Indeed, the communication literature suggests that behavioral cues are more ambiguous than verbal messages (Le Poire et al., 2002). Thus, avoidantly attached individuals may be able to enjoy the benefits of nonverbal affectionate communication, without needing to explicitly acknowledge the level of intimacy these types of behaviors convey.

Contributions to Literature on the Buffering of Attachment Avoidance

The present research makes a novel contribution to the literature on the buffering of attachment avoidance. This study is one of the first to specifically focus on effective strategies for communicating affection to avoidantly attached individuals. While there is a small body of research on how to tailor communication styles to insecurely attached individuals, most of it has focused on communication during more negative relationship issues such as conflict and asking a partner to change (Overall et al., 2013; Simpson, Winterheld, Rholes, & Oriña, 2007). The studies that have examined more positive aspects of communication in relationships have focused on the impact of self-disclosing positive information about the self (e.g., Gable et al., 2004; Hicks & Diamond, 2008); however, these studies have focused more on the benefits of these processes to the discloser rather than how the message can be most effectively communicated to the listener.

This study also provides novel insights into attachment theory by identifying the specific affectionate cues to which insecurely attached individuals may respond most strongly. For example, the present research is one of the few studies showing how highly avoidant individuals can respond to intimacy cues at levels similar to individuals low in attachment avoidance (Park et al., 2019; Stanton et al., 2017). While previous research has demonstrated that intimacy-related cues may broadly be associated with positive outcomes for avoidantly attached individuals, our research has identified which specific types of intimacy-laden cues may be necessary to achieve this effect, with nonverbal associated with particularly strong outcomes

Methodological Strengths, Limitations, and Future Directions

A major methodological strength of this work is its reliance on observational research methods to capture how best to communicate affection in a large sample of real-time couple interactions. Indeed, observational research provides an opportunity for naturalistic indicators of relationship behavior (Roberts & Greenberg, 2002). Furthermore, coded conversation behaviors by neutral third-party observers provide a less biased account of communication patterns than self-report measures of affectionate behaviors. Relatively few studies have utilized this technique to examine dyadic interactions, especially in affectionate exchanges (Roberts & Greenberg, 2002). Using observational approaches to studying individual differences in the receipt of affectionate cues can provide insight into which strategies may be most likely to evoke positive outcomes for those for whom intimacy and connection are difficult to achieve.

Although the observational approach to examining romantic relationships is a notable strength of the work, as with any study, there are some limitations to address. The first is that because this research is correlational, we cannot determine the direction of causality. Although it is possible that tailoring levels of nonverbal affection may have benefits for avoidantly attached individuals, it may also be the case that “happier” avoidantly attached individuals lead partners to feel more comfortable openly communicating love. However, because many of the patterns held when controlling for baseline emotions and relationship quality, we believe the former interpretation is more viable.

Second, because the research design was observational rather than experimental in nature, we cannot ascertain whether deliberate and intentional use of the communication patterns would have a positive impact on the listeners’ emotions. It is possible that the effects of nonverbal affection are only beneficial to the listener when they spontaneously arise in the conversation versus when partners are instructed to implement these strategies. Similarly, it is possible that the pattern of effects emerged because verbal affection was explicitly directed in our instructions to the participants, whereas nonverbal affection was not. Thus, because verbal affection was a necessary component of the task, avoidantly attached individuals may have been suspicious of its authenticity, thus undermining its potential impact on the listener. As such, if participants were directed to be nonverbally affectionate or enthusiastic with their partner, the effects of each communication style on avoidantly attached individuals’ outcomes may differ.

As a future direction, researchers could utilize an experimental design to explicitly coach people to use verbal affection and nonverbal affection techniques with their partners to examine if there is a causal impact of these communication patterns on the recipient’s positive relational outcomes. Indeed, knowing if certain patterns cause positive reactions in avoidantly attached individuals would inform whether verbal affection and nonverbal affection can be recommended to partners of avoidantly attached individuals as strategies to communicate feelings of love.

Relatedly, this research would benefit from examining the role of verbal affection and nonverbal affection in communicating love to avoidantly individuals over time. Although avoidant individuals responded positively to messages of nonverbal affection in the lab, it is possible that they could either become indifferent to—or overwhelmed by—such communication when used more chronically. Longitudinal data would strengthen the argument that verbal affection and nonverbal affection patterns are effective love communication strategies, and are needed before we can make confident recommendations regarding the application of these techniques to everyday relationship functioning.

Finally, another avenue for future research is to examine the mechanisms behind the unique benefits of nonverbal affection for avoidantly attached individuals. We have proposed that avoidantly attached individuals respond particularly well to nonverbal affection cues because these types of cues signal their partners’ reliability, trustworthiness, and willingness to care for their needs. Future research could more directly test these as mechanisms that drive avoidantly attached individuals’ well-being.

Conclusion

We have demonstrated in the present research that avoidantly attached individuals can benefit from affectionate exchanges with their romantic partners. However, those benefits may be somewhat conditional on how that affection is communicated. While partners’ nonverbal affection cues allow avoidantly attached individuals to experience positive behaviors to the same degree as those low in attachment avoidance, they may still also benefit from verbal affection to limited degrees, at least to keep negative emotions low.

Ultimately, tailored approaches to meeting avoidantly attached individuals’ unique needs (i.e., providing enhanced evidence of their partner’s trustworthiness) may allow them to lower their defenses and experience connection with more confidence in their partners’ affections, ideally leading to long-term reductions in attachment avoidance.

Supplemental Material

Schrage_Online_Appendix – Supplemental material for Effects of Verbal and Nonverbal Communication of Affection on Avoidantly Attached Partners’ Emotions and Message Receptiveness

Supplemental material, Schrage_Online_Appendix for Effects of Verbal and Nonverbal Communication of Affection on Avoidantly Attached Partners’ Emotions and Message Receptiveness by Kristina M. Schrage, Jessica A. Maxwell, Emily A. Impett, Dacher Keltner and Geoff MacDonald in Personality and Social Psychology Bulletin

Supplemental Material

Supplemental_Materials_revised_2 – Supplemental material for Effects of Verbal and Nonverbal Communication of Affection on Avoidantly Attached Partners’ Emotions and Message Receptiveness

Supplemental material, Supplemental_Materials_revised_2 for Effects of Verbal and Nonverbal Communication of Affection on Avoidantly Attached Partners’ Emotions and Message Receptiveness by Kristina M. Schrage, Jessica A. Maxwell, Emily A. Impett, Dacher Keltner and Geoff MacDonald in Personality and Social Psychology Bulletin

Footnotes

Authors’ Note

Jessica A. Maxwell is now affiliated with the University of Auckland, New Zealand.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by a Social Sciences and Humanities Research Council (SSHRC) Joseph-Armand Bombardier Canada Graduate Doctoral Scholarship to K. M. Schrage, a SSHRC Postdoctoral Fellowship awarded to J. A. Maxwell, a SSHRC Insight Grant awarded to G. MacDonald, a SSHRC Insight Grant awarded to E. A. Impett and a National Institute of Health Grant awarded to D. Keltner.

ORCID iD

Kristina M. Schrage

Notes

Supplemental Material

Supplemental material is available online with this article.

References

Andersen

P. A.

Guerrero

L. K.

Jones

S. M.

(2006). Nonverbal behavior in intimate interactions and intimate relationships. In Patterson

M. L.

(Ed.), The SAGE handbook of nonverbal communication (pp. 259–278). SAGE. https://doi.org/10.4135/9781412976152.n14

Booth-Butterfield

Trotta

M. R.

(1994). Attributional patterns for expressions of love. Communication Reports, 7(2), 119–129.

Brennan

K. A.

Clark

C. L.

Shaver

P. R.

(1998). Self-report measurement of adult romantic attachment: An integrative overview. In Simpson

J. A.

Rholes

W. S.

(Eds.), Attachment theory and close relationships (pp. 46–76). Guilford Press.

Coker

D. A.

Burgoon

J. K.

(1987). The nature of conversational nonverbal affection and nonverbal encoding patterns. Human Communication Research, 13, 463–494. https://doi.org/10.1111/j.1468-2958.1987.tb00115.x

Collins

N. L.

Feeney

B. C.

(2004). Working models of attachment shape perceptions of social support: Evidence from experimental and observational studies. Journal of Personality and Social Psychology, 87, 363–383. https://doi.org/10.1037/0022-3514.87.3.363

Comrey

A. L.

Lee

H. B.

(1992). A first course in factor analysis (2nd ed.). Lawrence Erlbaum.

Costello

A. B.

Osborne

J. W.

(2005). Best practices in exploratory factor analysis: Four recommendations for getting the most from your analysis. Practical Assessment, Research & Evaluation, 10, Article 7.

Curran

P. J.

Hussong

A. M.

(2009). Integrative data analysis: The simultaneous analysis of multiple data sets. Psychological Methods, 14(2), 81–100.

Farrell

A. K.

Simpson

J. A.

Overall

N. C.

Shallcross

S. L.

(2016). Buffering the responses of avoidantly attached romantic partners in strain test situations. Journal of Family Psychology, 30, 580–591. https://doi.org/10.1037/fam0000186

10.

Feeney

J. A.

(1999). Adult attachment, emotional control, and marital satisfaction. Personal Relationships, 6, 169–185.

11.

Floyd

(1997). Communicating affection in dyadic relationships: An assessment of behavior and expectancies. Communication Quarterly, 45(1), 68–80.

12.

Floyd

(2018). Affectionate communication in close relationships. Cambridge University Press.

13.

Fraley

R. C.

Waller

N. G.

Brennan

K. A.

(2000). An item response theory analysis of self report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365.

14.

Gable

S. L.

Reis

H. T.

Impett

E. A.

Asher

E. R.

(2004). What do you do when things go right? The intrapersonal and interpersonal benefits of sharing positive events. Journal of Personality and Social Psychology, 87, 228–245. https://doi.org/10.1037/0022-3514.87.2.228

15.

Girme

Y. U.

Overall

N. C.

Simpson

J. A.

Fletcher

G. J.

(2015). “All or nothing”: Attachment avoidance and the curvilinear effects of partner support. Journal of Personality and Social Psychology, 108, 450–475. https://doi.org/10.1037/a0038866

16.

Gosnell

C. L.

Gable

S. L.

(2013). Attachment and capitalizing on positive events. Attachment & Human Development, 15, 281–302. https://doi.org/10.1080/14616734.2013.782655

17.

Gupta

(2013). Effective body language in organizations. IUP Journal of Soft Skills, 7, 35–44.

18.

Hazan

Shaver

(1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52, 511–524. https://doi.org/10.1037/0022-3514.52.3.511

19.

Hicks

A. M.

Diamond

L. M.

(2008). How was your day? Couples’ affect when telling and hearing daily events. Personal Relationships, 15, 205–228. https://doi.org/10.1111/j.1475-6811.2008.00194.x

20.

Impett

E. A.

Gordon

A. M.

Kogan

Oveis

Gable

S. L.

Keltner

(2010). Moving toward more perfect unions: Daily and long-term consequences of approach and avoidance goals in romantic relationships. Journal of Personality and Social Psychology, 99, 948–963.

21.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). The analysis of dyadic data. Guilford Press.

22.

Le Poire

Duggan

Shepard

Burgoon

. (2002). Relational messages associated with nonverbal involvement, pleasantness, and expressiveness in romantic couples. Communication Research Reports, 19, 195–206. https://doi.org/10.1080/08824090209384848

23.

Medway

F. J.

Davis

K. E.

Cafferty

T. P.

Chappell

K. D.

(1995). Family disruption and adult attachment correlates of spouse and child reactions to separation and reunion due to Operation Desert Storm. Journal of Social and Clinical Psychology, 14, 97–118.

24.

Mikulincer

Nachshon

(1991). Attachment styles and patterns of self-disclosure. Journal of Personality and Social Psychology, 61, 321–331.

25.

Mikulincer

Shaver

P. R.

(2005). Attachment theory and emotions in close relationships: Exploring the attachment-related dynamics of emotional reactions to relational events. Personal Relationships, 12(2), 149–168.

26.

Mikulincer

Shaver

P. R.

(2007). Contributions of attachment theory and research to motivation science. In Shah

Gardner

(Eds.), Handbook of motivation science (pp. 201–216). Guilford Press.

27.

Mitchell

S. K.

(1979). Interobserver agreement, reliability, and generalizability of data collected in observational studies. Psychological Bulletin, 86, 376–390.

28.

Overall

N. C.

Simpson

J. A.

Struthers

(2013). Buffering attachment-related avoidance: Softening emotional and behavioral defenses during conflict discussions. Journal of Personality and Social Psychology, 104, 854–871. https://doi.org/10.1037/a0031798

29.

Park

Impett

E. A.

MacDonald

Lemay

E. P.

(2019). Saying “thank you”: Partners’ expressions of gratitude protect relationship satisfaction and commitment from the harmful effects of attachment insecurity. Journal of Personality and Social Psychology, 117, 773–806.

30.

Pistole

M. C.

(1994). Adult attachment styles: Some thoughts on closeness-distance struggles. Family Process, 33, 147–159. https://doi.org/10.1111/j.1545-5300.1994.00147.x

31.

Roberts

L. J.

Greenberg

D. R.

(2002). Observational “windows” to intimacy processes in marriage. In Noller

Feeney

J. A.

(Eds.), Advances in personal relationships. Understanding marriage: Developments in the study of couple interaction (pp. 118–149). Cambridge University Press. https://doi.org/10.1017/CBO9780511500077.008

32.

Rosenthal

Rosnow

R. L.

(2007). Essentials of behavioral research: Methods and data analysis (3rd ed.). McGraw-Hill.

33.

Rusbult

C. E.

Martz

J. M.

Agnew

C. R.

(1998). The investment model scale: Measuring commitment level, satisfaction level, quality of alternatives, and investment size. Personal Relationships, 5(4), 357–387.

34.

Shallcross

S. L.

Howland

Bemis

Simpson

J. A.

Frazier

(2011). Not “capitalizing” on social capitalization interactions: The role of attachment insecurity. Journal of Family Psychology, 25(1), 77–85.

35.

Shapiro

A. F.

Gottman

J. M.

(2004). The specific affect coding system. In Kerig

P. K.

Baucom

D. H.

(Eds.), Couple observational coding systems (pp. 205–222). Routledge.

36.

Simpson

Overall

(2014). Partner buffering of attachment insecurity. Current Directions in Psychological Science, 23(1), 54–59. https://doi.org/10.1177/0963721413510933

37.

Simpson

J. A.

Winterheld

H. A.

Rholes

W. S.

Oriña

M. M.

(2007). Working models of attachment and reactions to different forms of caregiving from romantic partners. Journal of Personality and Social Psychology, 93, 466–477. doi: 10.1037/0022-3514.93.3.466

38.

Stanton

S. C. E.

Campbell

Pink

J. C.

(2017). Benefits of positive relationship experiences for avoidantly attached individuals. Journal of Personality and Social Psychology, 113, 568–588. https://doi.org/10.1037/pspi0000098

39.

Tucker

J. S.

Anders

S. L.

(1998). Adult attachment style and nonverbal closeness in dating couples. Journal of Nonverbal Behaviour, 22, 109–124. https://doi.org/10.1023/A:1022980231204

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB

0.21 MB