Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 May 25.
Published in final edited form as: Motiv Emot. 2010 Jun 1;34(2):133–143. doi: 10.1007/s11031-010-9162-0

Composition and consistency of the desired affective state: The role of personality and motivation

Adam A Augustine 1, Scott H Hemenover 2, Randy J Larsen 3, Tirza E Shulman 4
PMCID: PMC3101480  NIHMSID: NIHMS296841  PMID: 21625402

Abstract

Using longitudinal and experience sampling designs, the consistency and composition, and personality and motivational predictors, of the desired affective state are explored. Findings indicate that, while the desired affect is relatively malleable throughout one semester, it is relatively stable throughout 1 week. Personality and motivations/goals were related to the content of the desired affective state. Extraversion, neuroticism, agreeableness, and conscientiousness were related to the content of the desired affective state. In addition, higher-order goals predicted the content of the desired affective state. Our results suggest that the content of the desired affective state may be largely dependent on personality, motivation, and, potentially, an interaction between personality and motivation.

Keywords: Desired affect, Ideal affect, Big 5, Motivation, Goals

To travel from Point A to Point B, you must first know the location of Point B. This fairly obvious statement has implications for all research on self-regulation, and particularly for research on affect regulation. An ever increasing body of literature focuses on affect regulation, or “processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions” (Gross 1998, p. 275). If individuals wish to influence what, when, and how they experience their emotions, they must have some goal for said influence. It is this goal, the desired affective state, which is the topic of the current research. In one longitudinal and one experience sampling we explore the consistency and composition of the desired affective state, as well as personality and motivational factors that may predict the content of one’s desired affective state.

Models of affect regulation and affect

The desired affective state (DAS) is a primary component of several of the prominent models of affect regulation that have surfaced in the extant literature. In both Bonanno’s (2001) and Larsen’s (2000) model of affect regulation, the goal of the affect-regulatory process is to reach the DAS. Individuals are aware of their DAS and go through a continual process in which they attempt to limit inconsistencies between their DAS and their current affective state. In Gross’ (1998) process model of emotion regulation, the importance of the DAS is more implicit. Two components of this model (situation selection and situation modification) require knowledge of the goal of the regulatory process. If one is to approach/avoid certain situations (situation selection) based on affective consequences or modify a situations emotional aspects/consequences (situation modification), one must have a goal (desired affect) for any approach or avoidance. Despite the centrality of the DAS to the affect-regulatory process, the extant literature yields relatively few studies on this construct. Those studies that do exist have largely focused on individual differences in the DAS as organized by the circumplex model of affect (see Larsen and Diener 1985, 1992; Russell 1980; Rusting and Larsen 1995).

In the circumplex model of affect, self-reported affect is organized around two dimensions, pleasantness and activation. Pleasantness represents a continuum from positive to negative and activation a continuum from high activation/arousal to low activation/arousal. When self-reported affect adjectives are plotted, they generally fall in a circular pattern organized around these two primary dimensions. When grouped based on their location in Cartesian space, eight types of affect adjectives emerge: high activation (i.e., surprised), low activation (i.e., idle), activated pleasant (i.e., elated), pleasant (i.e., glad), unactivated pleasant (i.e., content), activated unpleasant (i.e., fearful), unpleasant (i.e., miserable), and unactivated unpleasant (i.e., dull). Research regarding the DAS has yielded individual differences in both the desired activation and valence of affect.

The desired affective state

The desired affective state may depend on one’s personality. Given the strong and consistent relationships observed between extraversion and positive affect, and neuroticism and negative affect (for a review, see Larsen and Augustine 2008), one would expect that these personality traits would predict the DAS. Indeed, Rusting and Larsen (1995) found that the level of affect activation one typically experiences is related to the amount of affect activation that one desires, although the direction of this relationship is dependent on personality. Extraverts (vs. introverts) experience more activated emotions and wish to experience more activated emotions, be they pleasant or unpleasant. Neurotics (vs. emotionally stable) experience more activated negative affects but wish to experience affects that are less activated. In contrast to these findings, Kampfe and Mitte (2009) found that neuroticism positively predicted a desire for unactivated unpleasant affects and negatively predicted desires for activated pleasant affect; this pattern is more consistent with the actual affective experience of the neurotic. In line with these findings for neuroticism, Tamir (2005) found that those high in neuroticism desire trait-consistent negative affect (i.e., higher negative affect) when driven by performance goals.

Stable and short-term motives and goals have also been linked to the content of the DAS. Interpersonal goals predict the DAS both within and across cultures (Tsai et al. 2007). Individuals and cultures (i.e., European Americans) that value influencing others (making others suit one’s personal needs) desire more activated pleasant affect. Individuals and cultures (i.e., those from Hong Kong) that value adjusting to others (making yourself suit other’s needs) desire more unactivated pleasant affect. Additionally, it was found that endorsement of these interpersonal goals mediated cultural differences in the DAS. Individuals’ affective desires also depend on whether affect is to be used or merely experienced. When under a utilitarian (using affect) motivation, the content of the DAS may be different than when operating under a hedonic (experience) motivation (Tamir et al. 2007). While neurotics may be generally motivated (under normal hedonic motivations) to experience affects that are inconsistent with trait levels (Rusting and Larsen 1995), those high in neuroticism are motivated to experience trait-consistent affect when operating under a utilitarian motivation. In this case, achieving trait-consistent affect is beneficial for performance (Tamir 2005). In addition, affective desires depend on immediate motivations to achieve rewards or avoid threats. When motivated to achieve rewards, individuals desire approach related affects such as excitement or happiness. However, when motivated to avoid threats, individuals desire avoidance related affects such as worry or fear (Tamir et al. 2007). Thus, the content of the DAS may vary based on long-term, short-term, interpersonal, hedonic, or pragmatic (i.e., performance) motivations and goals.

In sum, the content of the desired affective state may be shaped by two1 factors: personality and motivation. In the current research, we explore the roles that personality, motivation, and their interaction, play in determining the content of the DAS.

Overview of the current research

The current research is aimed to examine the composition and consistency of the desired affective state. Two primary hypotheses guide the current research. First, personality is related to the content of the DAS. Second, motives/goals will predict the content of the DAS. Study 1 addresses the first primary hypothesis by measuring the DAS and the Big Five personality traits. Given prior research (Rusting and Larsen 1995), we hypothesize that extraversion and neuroticism will predict desired activation levels, with extraverts desiring trait-consistent and neurotics trait-inconsistent affect activation. Conscientiousness and agreeableness both predict experienced positive affect (Costa and McCrae 1980; McCrae and Costa 1991; Watson and Clark 1992). These two traits also predict self-regulation behavior, with conscientiousness predicting an increased use of effective primary coping2 strategies (Connor-Smith and Flachsbart 2007) and agreeableness predicting increased efforts to control and regulate affect (Tobin et al. 2000; Meier et al. 2006). Those high (vs. low) in conscientiousness and agreeableness seem to aim their regulatory efforts at maintaining a pleasant valence of affect, and are successful in their attempts. Given the links between these two traits and an increased frequency of apparently successful efforts to achieve pleasant affect, we hypothesize that conscientiousness and agreeableness, will predict a desire for the valence of affective states; those high in these constructs will desire affective states with a more pleasant tone. In other words, if those higher in agreeableness and conscientiousness put a great deal of effort into achieving pleasant states, then this may be indicative of a desire for pleasant states. Although openness is related to affect (such that those higher in openness experience more pleasant affects), this relationship may be largely due to reactions to the varied and complex experiences of the person high in openness (i.e., McCrae 2007), rather than affect regulation behavior. As such, we do not expect openness to predict the DAS.

To address the second primary hypothesis, participants in Study 1 will provide free-response descriptions of their motivations for choosing each specific DAS. Although Tsai et al. (2007) found that the DAS differed across two interpersonal goals (influence of and adjustment to others), the relation of the DAS to other goals is unknown. Given that experienced affect is related to higher-level motivations/goals (i.e., power, achievement, etc.; Nelissen et al. 2007), desired affect may be linked with higher-level goals. Using this free response data gathered from Study 1, a set of seven higher-level motives/goals for the DAS will be derived. Using an experience sampling design, in Study 2 we explore differences in the DAS based on these seven higher-level goals. Study 2 also contains a conceptual replication of Study 1 and, in addition, allows us to explore the role of the DAS in the affect-regulatory process by measuring affect regulation style.

Finally, both studies will test the temporal consistency of the DAS, which has been examined only once in prior research (mean test–retest r = .61; Tsai, Knutson, and Fung 06). To test this, participants will make multiple ratings of their DAS.

Study 1

Method

Participants

Participants were drawn from general psychology courses at a large Midwestern university and received course credit for participation. This study utilized a sample of 200 participants (due to IRB concerns regarding confidentiality, participant demographics were not gathered) and yielded 773 reports.

Materials: Questionnaires

Personality

The Big Five personality traits were assessed using the BFI-54 (Big Five Inventory; John et al. 1991) which utilizes a 5 point, disagree strongly—agree strongly, Likert-type scale (in this study: extraversion α = .87, agreeableness α = .81, conscientiousness α = .74, neuroticism α = .77, openness α = .83).

Desired affective state

The content of the participants’ DAS was measured using 62 affect adjectives taken from the PANAS (Watson et al. 1988) and the circumplex model of affect (Rusting and Larsen 1995). The combination of these two scales yields ten different subscales (positive affect, PA, mean α = .88; negative affect, NA, mean α = .89; high activation, HA, mean α = .71; activated pleasant, AP, mean α = .79; pleasant, P, mean α = .86; unactivated pleasant, UP, mean α = .74; low activation, LA, mean α = .72; unactivated unpleasant, UU, mean α = .91; unpleasant, U, mean α = .91; activated unpleasant, AU, mean α = .78). Participants indicated to what extent they wished to experience each of the 62 affects using a 5-point, very slightly or not at all—extremely, Likert-type scale.

For ratings of the DAS, participants were told to indicate (using the adjective scales) how they would like to feel and were instructed as follows, “This part of the questionnaire deals with how you would like to feel most of the time. Please mark the extent to which each word describes how you would feel on a perfect day, that is your ideal emotional state.” Participants also described (using free-response) how they wanted to feel on a perfect emotional day and why they wanted to feel that way. Unfortunately, a large number of responses to the first question were not regarding emotional experience (i.e., I want to play basketball all day). While this made the answers to the two questions largely unusable, responses to the second question were used to generate a questionnaire for Study 2.

Procedure

Participants completed the DAS questionnaires on four separate occasions; each report was made approximately 25 days after the prior report. Questionnaires were administered online; all participants expressing interest in participating in the study had access to the internet. Announcements as to when each testing session would be available were made in person (during psychology courses) and through e-mail. After making each testing session available online, participants were given 1 week to complete the procedure.

During the first session (T1), participants completed the BFI-54 and rated their DAS. At Times 2–4, participants completed the DAS questionnaire. Additionally, during each session participants were asked, using a free-response format, how they wanted to feel on the perfect emotional day and why that wanted to feel that way. The responses to these questions were not used to calculate participants’ DAS (this data was gathered using the DAS questionnaire described previously), but were used for the generation of a questionnaire for Study 2.

Results & discussion

Stability

To examine the stability of the general DAS, mean test–retest correlations and unconditional intraclass correlations were calculated for each of the 10 affect subscales (see Table 1). Test–retest correlations indicate moderate stability for the DAS (mean r = .47–.59; T1, T2, T3, and T4). Thus, the DAS is somewhat malleable (relative to more stable personality variables) over the course of a semester. In addition, roughly equal amounts of variance (ICC r = .44–.56) existed in the DAS at the between and within-person levels.

Table 1.

Studies 1 & 2: Mean DAS data

General (S1)
 General (S2)
M SD r (ICC r) M SD r (ICC r)
Positive affect (PA) 3.85 .55 .59 (.55) 3.87 .55 .77 (.71)
Negative affect (NA) 1.48 .46 .54 (.53) 1.13 .24 .59 (.50)
High activation (HA) 3.15 .53 .54 (.52) 2.96 .59 .77 (.73)
Activated pleasant (AP) 3.53 .61 .58 (.56) 3.67 .64 .76 (.72)
Pleasant (P) 4.10 .56 .54 (.52) 4.30 .46 .64 (.47)
Unactivated pleasant (UP) 3.55 .52 .48 (.48) 3.85 .62 .68 (.60)
Low activation (LA) 2.24 .54 .58 (.55) 2.05 .53 .77 (.69)
Unactivated unpleasant (UU) 1.49 .54 .50 (.48) 1.16 .30 .68 (.52)
Unpleasant (U) 1.34 .47 .47 (.44) 1.10 .24 .60 (.43)
Activated unpleasant (AU) 1.72 .52 .56 (.55) 1.18 .29 .63 (.56)
Open in a new tab

Note: S1 = Study 1, S2 = Study 2; N = 170 (S1), N = 125 (S2); r indicates mean test–retest correlation, ICC r indicates the unconditional intraclass correlation; all other data represents mean responses across time

Relationship to personality

Given the hierarchical structure of the data set (reports nested within participants) hierarchical linear modeling was used to examine our predictions (Raudenbush and Bryk 2002). All affective states were examined in separate, intercept-only models. At level 1, the DAS was represented with the equation: yij = β0j + rij. At this level, yij represents the DAS for person j on occasion i, β0j represents the mean of y for person j across time, and rij residual variance. At level 2, we predicted the mean value of the DAS for each person using the equation: β0j = γ00 + γ01(Extraversion) + γ02(Agreeableness) + γ03(Conscientiousness) + γ04(Neuroticism) + γ05(Openness) + u0j. In level 2 equation, γ01 through γ05 represent the Big Five personality traits and u0j represents the error variance at the person level. A significant effect for any of the personality predictor variables indicates that the trait being examined is related to the mean of that particular DAS.

Consistent with our hypotheses (Rusting and Larsen 1995), extraverts (vs. introverts) desire affects in line with their typical experience; extraverts desire more activated affects (i.e., AP), even if those affects are negative (i.e., AU; see Table 2). In addition, neurotics desired trait inconsistent affect in the form of increased levels of pleasant affect (i.e., AP, P). Also consistent with our hypotheses, conscientiousness and agreeableness predict trait-consistent affective desires; those high in these constructs wish to experience more pleasant (i.e., AP, P) and less unpleasant affect (i.e., U, AU), regardless of activation level. Agreeableness is a trait of politeness and conscientiousness is a trait of organization (both personal and social; Costa and McCrae 1992). It may be that possessing a truly hedonic (higher pleasant and lower unpleasant, regardless of activation) affective profile facilitates maintenance of the polite and organized demeanors of these individuals high (vs. low) in agreeableness and conscientiousness. Finally, no associations between openness and the DAS were observed. Thus, consistent with experienced affective states and our hypotheses, extraversion predicts desired activation, agreeableness/conscientiousness predict desired pleasantness, and neuroticism predicts a desire for trait-inconsistent levels of positive affect.

Table 2.

Study 1: Relationship between personality and mean DAS ratings

PA NA HA AP P UP LA UU U AU
E .02 (.01)* .01 (.01)* .02 (.01)* .02 (.01)* .02 (.01)* −.01 (.01)* −.02 (.01)* .00 (.01) .01 (.01) .02 (.01)*
A .01 (.01) −.02 (.01)* −.01 (.01) .02 (.01) .03 (.01)* .02 (.01)* .00 (.01) −.02 (.01) −.02 (.01)* −.02 (.01)
C .03 (.01)* −.02 (.01)* .00 (.01) .01 (.01) .01 (.01) .00 (.01) −.02 (.01)* −.03 (.01)* −.02 (.01)* −.02 (.01)*
N .01 (.01) .00 (.01) .00 (.01) .02 (.01)* .02 (.01)* −.01 (.01) −.01 (.01) .00 (.01) .00 (.01) .00 (.01)
O .00 (.00) .00 (.00) .01 (.00) .00 (.00) .00 (.00) .00 (.00) .00 (.00) .00 (.00) .00 (.00) .00 (.00)
Open in a new tab

Note: Level 1, n = 773; level 2, n = 200; all results represented as unstandardized beta (standard error)

*

p < .05

In sum, Study 1 demonstrated that the DAS is related to the personality traits of extraversion, neuroticism, agreeableness, and conscientiousness. Additionally, the DAS is somewhat malleable over the course of one semester.

Study 2

The purpose of Study 2 is to replicate and extend the findings of Study 1 by investigating the DAS over the course of 1 week using an experience sampling method. First, to confirm the findings of Study 1, personality is again measured in this sample. We expect the same pattern of results to emerge in this study.

Second, to explore the role of the DAS in the affect-regulatory process, affect regulation style is measured. If the goal of the affect regulatory process is indeed the DAS, one would expect relations to emerge between affect regulation style and the DAS. Affect regulation strategies differ in their effectiveness for reaching certain affective states (Augustine and Hemenover 2009) and would, thus, be differentially effective for reaching certain DASs. If individuals were aware of the best strategies for reaching a given DAS, then they would be more likely to use certain strategies to achieve that state. Unfortunately, little research exists as to the efficacy of certain affect regulation strategies for the attainment of specific affects (i.e. high activation, activated pleasant, etc.); as such, a priori predictions as to the relationships between the use of certain affect regulation strategies and affective desires would be tentative at best.

Third, the consistency of the DAS will again be examined in this study. Although the results of Study 1 suggest only moderate stability, this study takes place over a shorter time period and, as shorter time periods lead to larger stability estimates, the DAS should exhibit stronger stability. Fourth, the free response data gathered as a part of Study 1 will be used to create a questionnaire assessing broad motivations for the DAS. Prior research has demonstrated that actual emotional experience is linked to broad motives/goals (Nelissen et al. 2007). Additionally, Tsai et al. (2007) demonstrated that the DAS was predicted by interpersonal motivations both within and across cultures. As such, we predict that the DAS will vary based on the broad motives/goals that were gathered (and subsequently used as a guide for scale construction) as part of Study 1. Differences in the content of the DAS, while under varying motivations, should be indicative of the motivation in question (i.e. those with a hedonic motive desire more pleasant affects, those with a relaxation motive desire more low activation affects, etc.).

Method

Participants

All participants were drawn from a psychology subject pool at a private Midwestern University and received course credit for participation. This study utilized a total sample of 156 participants (Age: M = 19.49, SD = 1.19; 20% male) who made a total of 809 reports.

Materials: Questionnaires

Personality

The Big 5 personality traits were assessed using the Five Factor Inventory (NEO-FFI: Costa and McCrae 1992). Responses on the NEO-FFI are made on a 5-point, strongly disagree—strongly agree, Likert-type scale (in this study: extraversion α = .72, agreeableness α = .74, conscientiousness α = .84, neuroticism α = .86, openness α = .78).

Affect regulation style

Participants’ affect regulation patterns were measured using the Measure of Affect Regulation Style (MARS: Larsen and Prizmic 2004). The MARS uses 38 items to assess how frequently individuals use certain affect regulation strategies. All responses on the MARS are made on a 7-point, not at all—almost always, Likert-type scale. The MARS yields seven subscales: active distraction (i.e., I go out with my friends; α = .68), cognitive engagement (i.e., I try to put things into perspective; α = .72), behavioral engagement (i.e., I take action to solve the problem; α = .64), venting (i.e., I let my feelings out by expressing them; α = .69), passive distraction (i.e., I eat something; α = .33), rumination and withdrawal (i.e., I withdraw from or avoid the situation; α = 35), and waiting and reframing (i.e., I do nothing; α = .44). Although the last three subscales yielded low internal consistency, this is in line with prior work using the MARS; these subscales contain only 4 items each with rather low inter-item relationships.

Desired affective state

Measurement of the DAS in this study was identical to Study 1. All subscales achieved adequate internal consistency (PA mean α = .88; NA mean α = .92; HA mean α = .78; AP mean α = .87; P mean α = .68; UP mean α = .73; LA mean α = .79; UU mean α = .91; U mean α = .92; AU mean α = .86).

Motives for desired affective state

As part of Study 1, participants responded to two open ended questions regarding their DAS: how do you want to feel on your perfect emotional day and why do you want to feel that way. While a multitude of off-topic answers to the first question did not allow data analysis, responses to the second question (why do you want to feel that way) were used to create a questionnaire (using an informal act-frequency approach) to measure motives for the DAS. The motives used in this questionnaire are perfect balance (i.e., this gives me the perfect balance of emotions), hedonic (i.e., I want to be happy, feel good, enjoy myself, etc.), productivity (i.e., I want to get things done, be productive, etc.), relaxation (i.e., I want to be relaxed, care-free, have no worries, have time to think, etc.), avoiding anhedonic (i.e., I don’t want to feel bad, nervous, anxious, sad, angry, etc.), self-verification (i.e., I normally feel this way and this makes me feel like myself, natural, etc.), and sociability (i.e., This makes it easier to be around people, makes me personable, etc.).

Participants were asked to select one of seven motives that most closely matched their reasons for choosing their DAS on that particular occasion. In case the seven items did not fully capture motives, an “other” selection was offered. This “other” selection was chosen on 16 (of 809 total) occasions. For 15 of these, participants mentioned two motives; in these cases the first motive mentioned was entered as the participant’s response (the other case was deleted).

In addition to providing a measure of which motive was most important in participants’ affective desires, this also provides a measure of motive variability. The participants motive variance, or the frequency of motive changes over the course of the reporting period, was calculated by counting the number of times each participant’s motive changed from the previous report and dividing that number by the total number of reports completed.

Procedure

Participants completed questionnaires regarding their DAS everyday for a period of 7 days. The questionnaires were administered online; all participants expressing interest in participating in the study had access to the internet. For the first days report, participants were brought into the lab in groups of five and completed the NEO-FFI, MARS, and the first DAS and motive reports. Participants were then debriefed and released to complete each day’s DAS and motive reports on their own for the next 6 days.

During the session in the lab, each participant was given a schedule instructing them on when to complete each daily report; participants were randomly assigned to complete each day’s report either between 8 a.m. and 1 p.m., 1 p.m. and 6 p.m., or 6 p.m. and 11 p.m. The online survey tool reported the time and data on which participants completed each report. Approximately 86% of the daily reports were made during the assigned time period. Given the high compliance rate, all reports from these participants are used in the full data analysis.

Results & discussion

Stability

Mean test–retest correlations (mean r = .59–.77) and unconditional intraclass correlations (r = .47–.73) were calculated for each of the ten affect subscales and can be viewed in Table 1. Results indicate that, consistent with the shorter time period involved and our predictions, the DAS is relatively stable (at a mean level) over the course of 1 week. In addition, while more variation exists between individuals than within-individuals for positive affects, the amount of between and within-person variance for activated pleasant affect, unactivated pleasant affect, high activation affect, and low activation affect is roughly equivalent.

Personality

The same multi-level equations described in Study 1 were used to examine the relationships between the DAS and the Big Five (see Table 3). For examinations of affect regulation style, each affective state was again examined in separate, intercept-only models. These models utilized the same level 1 equations described previously. However, at level 2, each affect regulation style component was examined separately in the equation: β0j = γ00 + γ01(Trait) + u0j. In level 2 equation, γ01 represents the affect regulation style component being examined and u0j represents the error variance at the person level (see Table 3).

Table 3.

Study 2: Relationship between personality and the DAS

PA NA HA AP P UP LA UU U AU
E .42 (.11)* .13 (.06)* .33 (.10)* .63 (.11)* .38 (.09)* −.24 (.14) −.26 (.12)* .06 (.06) .10 (.06) .15 (.07)*
A .25 (.08)* −.10 (.05)* .00 (.10) .31 (.10)* .35 (.07)* .20 (.11) −.05 (.08) −.10 (.05) −.09 (.05) −.10 (.06)
C −.03 (.10) −.03 (.05) −.13 (.10) .05 (.11) −.07 (.09) .04 (.10) .14 (.09) .05 (.06) −.01 (.05) −.06 (.06)
N .17 (.07)* .04 (.03) .03 (.08) .26 (.09)* .18 (.05)* .04 (.09) −.08 (.08) −.02 (.03) .05 (.04) .03 (.04)
O .00 (.09) −.03 (.05) .13 (.10) .00 (.10) .01 (.08) .07 (.10) .00 (.08) −.01 (.06) .00 (.05) −.04 (.06)
Active distraction .20 (.07)* .07 (.04) .25 (.07)* .31 (.08)* .20 (.07)* .02 (.07) −.06 (.06) .08 (.04)* .06 (.04) .09 (.04)*
Cognitive engagement .10 (.06) .05 (.02)* .11 (.06) .09 (.07)* .11 (.05)* .06 (.06) .05 (.04) .06 (.03)* .07 (.02)* .07 (.03)*
Behavioral engagement .23 (.07)* .07 (.03)* .23 (.07)* .29 (.07)* .20 (.06)* .02 (.07) .01 (.06) .09 (.03)* .07 (.03)* .07 (.04)
Venting .09 (.05)* −.01 (.02) .04 (.05) .17 (.06)* .15 (.04)* −.02 (.05) −.08 (.04)* −.02 (.02) .00 (.02) .00 (.03)
Passive distraction .03 (.04) .05 (.02)* .09 (.05) .07 (.05)* .09 (.04)* .07 (.05) .08 (.04) .04 (.02) .05 (.02)* .06 (.03)*
Rumination .00 (.06) .02 (.02) .02 (.07) .02 (.07) .12 (.05)* .13 (.06)* .04 (.06) .01 (.02) .04 (.03) .02 (.03)
Waiting −.02 (.06) .04 (.03) .03 (.05) .03 (.06) .07 (.04) .06 (.06) .09 (.05) .04 (.03) .06 (.03) .06 (.03)
Open in a new tab

Note: Level 2, n = 156; Level 1, n = 809; all effects represented as unstandardized beta (standard error)

*

p < .05

The findings for personality were generally consistent with those of Study 1 and the first primary hypothesis for these studies; personality predicts the DAS. Extraverts (vs. introverts) desire more activated affects. Those higher in neuroticism desire more pleasant affects. Those high (vs. low) in agreeableness, desire more pleasant affects. However, in this study, no associations were observed between agreeableness and the desire for unpleasant affects. As in Study 1, no associations between the DAS and openness were observed. Finally, no significant effects emerged for conscientiousness; the reasons for this disconnect between Studies 1 and 2 are unclear.

If the goal of the affect regulatory process is indeed the DAS, one would expect relations to emerge between affect regulation style and the DAS. In line with this, individual’s typical affect regulation style was related to their DAS (see Table 3). At a broad level (and consistent with models of affect regulation, i.e., Bonanno 2001; Larsen 2000) this suggests that individuals may shape their affect regulation style to reach their DAS. For instance, those who use cognitive engagement strategies more frequently desire higher levels of unpleasant affects (regardless of activation). Perhaps individuals who engage these emotions gain some benefit from that engagement. Thus, increased unpleasant affect would lead to an increase in the benefits of cognitive engagement.

Those who use venting more frequently desire higher levels of positive, activated pleasant, and pleasant affects. Given that the use of venting is associated with an increase in negative affect (Geen and Quanty 1977; Bushman 2002), these individuals may wish to experience something other than the increased negative affect that accompanies their typical affect regulation style. The frequent use of passive distraction predicts a desire for more activated affects, regardless of their valence. The use of these passive strategies may lead to a relatively unactivated emotional life, and thus, a potentially boring emotional life. These individuals may desire more activation to break up the monotony created by their typical regulation style. Finally, those who ruminate more frequently are more likely to desire pleasant affects. This is not surprising given the increase in unpleasant affects that would naturally follow from consistently focusing on all the bad things that happen in one’s life.

Motivation

To explore the relationship between motivation and the DAS, several analyses were conducted. First, to determine if the DAS varied over time as a function of motive, within-person standard deviations were calculated for each participant completing four or more reports (to ensure adequate calculation of variance). Motive variance (how much participants changed their desired affect motive) predicted intraindividual variance (controlling mean level relationship to eliminate artifacts) in both low activation affect and all positive affect subscales (PA, pr = .36, p < .05; NA, pr = .10, ns; HA, pr = .21, p < .05; AP, pr = .27, p < .05; P, pr = .24, p < .05; UP, pr = .23, p < .05; LA, pr = .36, p < .05; UU, pr = .10, ns; U, pr = .09, ns; AU, pr = .11, ns). This supports our hypothesis that the content of the DAS should be predicted by the motive for that state.

To further examine the relationship between the DAS and motives, a MANOVA was conducted (all ten affect subscales were entered as DVs, motive choice as IV). Results indicate that (see Table 4), for nine of the ten affect types (excluding unactivated unpleasant), motive did predict the DAS (Fs (6, 796) = 2.58–16.56, p < .05). Post-hoc analyses (using a Bonferroni adjustment) indicate that those with a perfect balance, hedonic, productivity, or self-verification motive desire more positive affect than those with a relaxation motive. Those with a perfect balance motive desire more negative affect than those with a hedonic motive. Those with a hedonic motive desire more high activation affect than those with a relaxation motive. Those with perfect balance, productivity, and relaxation motives desire less activated pleasant affect than those with a hedonic motive. Additionally, those with a self-verification motive desire more activated pleasant affect than those with a relaxation or perfect balance motive. Those with a hedonic motive desire more pleasant than those with a productivity or relaxation motive. Individuals with hedonic and productivity motives desire more unactivated pleasant affect than those with relaxation motives. Those with relaxation motives desire more low activation affect than those with perfect balance, hedonic, productivity, or avoid anhedonic motives. Additionally, those with perfect balance motives desire more low activation affect than those with self-verification motives. Those with perfect balance motives desire more unpleasant affect than those with hedonic motives. Finally, those with a productivity motive desire more activated unpleasant affect than those with a hedonic motive.

Table 4.

Study 2: DAS motives

Perfect balance Hedonic Productivity Relaxation Avoid anhedonic Self-verification Sociability
n 88 405 68 145 36 52 9
PA 3.73 (.80)a 3.98 (.65)b 4.01 (.71)b 3.58 (.67)a 3.64 (.65) 3.96 (.55)b 3.61 (.49)
NA 1.22 (.48)a 1.10 (.29)b 1.22 (.42) 1.10 (.28) 1.12 (.54) 1.12 (.28) 1.01 (.03)
HA 2.99 (.81) 3.02 (.72)a 3.03 (.72) 2.74 (.71)b 2.66 (.55) 2.85 (.53) 2.74 (.36)
AP 3.42 (.83)a 3.88 (.72)b 3.45 (.88)a 3.36 (.88)a 3.46 (.87)a 3.85 (.64) 3.20 (.79)
P 4.21 (.67) 4.43 (.72)a 4.12 (.70)b 4.17 (.66)b 4.13 (.72) 4.36 (.52) 4.03 (.56)
UP 3.77 (.72)a 3.82 (.87)a 3.56 (.67)a 4.29 (.58)b 3.98 (.69) 3.58 (.76)a 4.02 (.83)
LA 2.15 (.55)a 1.94 (.61)a 1.96 (.59)a 2.49 (.72)b 2.06 (.74)a 1.77 (.56)a 2.15 (.49)
UU 1.23 (.60) 1.12 (.38) 1.23 (.48) 1.21 (.38) 1.16 (.51) 1.11 (.34) 1.00 (.00)
U 1.23 (.57)a 1.07 (.29)b 1.15 (.40) 1.11 (.35) 1.11 (.61) 1.06 (.27) 1.01 (.06)
AU 1.27 (.49) 1.15 (.36)a 1.31 (.50)b 1.14 (.34) 1.13 (.52) 1.15 (.31) 1.02 (.06)
Open in a new tab

Note: n = 803 reports; differing superscripts letters indicate difference is significant at the p < .05 level

Thus, consistent with our predictions, the affects desired were in line with the DAS motive. This analysis also reveals that, in the majority of cases (50.4%) individuals’ primary motive for the DAS was a hedonic one. It seems that (consistent with Larsen’s (2000) set point model of affect regulation) individuals are usually operating under a general motivation to feel good (i.e., the hedonic hypothesis).

In sum, Study 2 confirmed the relationships between extraversion, agreeableness, neuroticism and the DAS that were observed in Study 1. Study 2 also found that affect regulation style was related to the content of the DAS. Consistent with our hypotheses, one’s motive for the DAS predicted the content of the DAS. Finally, it seems that individuals operate under a motivation to feel good the majority of the time.

General discussion

One of the most important parts of any regulatory system is the goal of regulation efforts. If one wishes to influence what, when, and how they experience their emotions, they must have a goal for that influence. The current research examined the goal of the affect-regulatory system, the desired affective state (DAS). In one longitudinal and one experience sampling study, the consistency and composition of the DAS was investigated. In addition, ways in which personality and motivations influence the DAS were explored. Findings indicate that, while the DAS is relatively malleable over the course of one semester, it is somewhat stable over the course of 1 week. Results also indicate that, while roughly equivalent variance is observed at the between and within-person levels over the course of one semester, this is only the case for negative affects over the course of 1 week. In Study 2, more variance was observed between-persons for pleasant affects and affects lacking a hedonic tone (i.e., HA and LA). This suggests that these affective states are relatively more stable within-person across a shorter time period.

Motivations and goals are related to the content of the DAS. In Study 2, the broad motivations for desired affect that were gathered as a part of Study 1 showed relationships with the DAS. Changes in these motives predicted variance in the DAS. In addition, observed differences in the content of the DAS between motives reflected the nature of those motives (i.e., individuals operating under a hedonic motive desired more pleasant affects). These findings suggest that the motivation for the DAS is important in determining the content of that state.

Both personality and affect regulation style were related to the content of the DAS.3 In both studies, extraversion, neuroticism, and agreeableness were related to the content of the general DAS. In addition, numerous relationships were observed between affect regulation style and the DAS. This suggests that individuals shape their affect regulation efforts in an effort to reach their DAS.

Implications

The findings of these studies, when taken in concert with the findings of Rusting and Larsen (1995), Tsai et al. (2007), Tamir (2005), and Tamir et al. (2007) suggest that personality and motivation may, to a large extent, determine the content of the DAS. Results from these studies, Rusting and Larsen (1995), and Tamir (2005) clearly indicate that personality is related to the content of the DAS. Extraversion predicts a desire for activated affect and agreeableness predicts a desire for positively valenced affect. Neuroticism predicts a desire for trait-inconsistent affect in general, but when under performance motivations, neuroticism predicts a desire for negative affect. Thus, as is the case with the affect regulatory process as a whole, the goal of affect regulation is also related to personality.

Motivations also predict the content of the DAS. Broad (Study 2), interpersonal (Tsai et al. 2007), and utilitarian (Tamir et al. 2007) motivations have all been shown to predict the content of the DAS. Indeed, Tsai et al. (2007) found that motivation mediated cultural variation in the DAS. Given this, the findings that broad motivations influence the DAS, findings that indicate that higher order values/motivations influence experienced affect (Nelissen et al. 2007), and findings that motives predict temporal variance in the DAS (Study 2), it seems clear that motives have an influence on desired affect.

Motivations and personality may also interact to shape the content of the DAS. In Study 2, we found that the DAS varied between broad level motives. Previous research suggests that the content of the DAS may depend not only on these motives, but also the type of person operating under these motives. Although individuals are typically motivated to feel good (i.e., hedonic motivation, Study 2), Tamir (2005) found that neurotics can be motivated to experience trait-consistent affect when engaging in demanding tasks. This relationship is contrary to the findings for neuroticism when not necessarily motivated to engage in any task (as in these studies and Rusting and Larsen 1995). These findings point to a person x motivation interaction, whereby certain individuals desire certain affects under certain motivations. Unfortunately, due to the structure of our data, we were unable to test for a statistical interaction between motives and personality. Additional studies, perhaps having participants engage in desired affect motivation perspective taking (i.e., asking them to give their DAS for a particular motivation), are required to further explore this issue.

These studies and prior research suggest that the DAS can be dependent on one’s actual affective experience. Past research on this construct has found that desired and experienced affect are relatively independent, although still highly related (actual—desired r = .28–.53 for pleasant affects; r = .11–.22 for unpleasant affects; r = .47 and .31 for low and high activation affects, respectively; Tsai et al. 2006). Given this, the effects of experienced affect have been controlled when analyzing cultural predictors of the DAS. However, in the current research, experienced affect was not entered as a covariate (nor was it measured). The results of these studies as well as several other lines of research support this choice.

First, some individuals should be relatively satisfied with their typical affective state. Individuals generally desire positive affect (Larsen 2000) and most individuals typically experience positive affect (Diener and Diener 1996). Thus, since people generally want to be happy and generally are happy, then the goal state and actual experience should be similar. Second, individuals are typically motivated to seek out experiences that confirm their selfconcepts (Swann et al. 1992). This desire to confirm one’s self-concept may also occur with affective experience. As affective experience has a stable component, trait affect may become part of one’s self-concept. As with other components of one’s self-concept, individuals would seek out experiences that promote the affective elements of their self-concept. Third, data from these studies, Rusting and Larsen (1995), and Tamir (2005) demonstrate that some individuals do wish to experience trait-consistent affect. Extraversion predicts the desire for a trait-consistent pattern of activation. Agreeableness and conscientiousness predict a desire for a trait-consistent pattern of pleasantness. Neuroticism predicts a desire for trait-consistent negative affect under performance motivations (Tamir 2005). The majority of the time, individuals are motivated to (and do, i.e., Diener an Diener 1996) experience affects in line with the hedonic hypothesis (i.e., Study 2; Larsen 2000). Thus, those individuals who, at a trait-level, experience relatively more pleasant affect would desire to continue that experience.

Finally, if actual affect were controlled in studies of the DAS, it is unclear exactly what construct one is examining. If an individual says that they want to be happy, then they want to be happy. If one controls the actual level of happiness for a happy person who wants to be happy, then their desire for happiness would be lost; the construct being examined is no longer a goal state, but a leftover. This new construct, with an actual affect covariate, may be more representative of another potentially important construct, regulatory success. With actual affect controlled, one is likely examining something that would represent the difference between actual and desired affect. While this difference, which is perhaps a measure of regulatory success, is certainly of great importance for the study of affect regulation behavior, it is not representative of a goal state. Recent research on the difference between actual and desired affect (affective discrepancy, Kampfe and Mitte 2009) supports this idea. Those higher in extraversion and lower in neuroticism show lower discrepancies between actual and desired affect (Kampfe and Mitte 2009), and this is consistent with the increased regulatory abilities of those individuals (Shulman et al. 2006; Hemenover et al. 2008). Thus, the extent to which certain personality and motivational constructs predict the DAS may depend on their relationships with actual affective experience and controlling actual affect degrades the ability to accurately examine desired affect. While experienced affect may not always match desired affect, there are cases wherein these two are roughly equivalent; generally happy people wish to remain generally happy.

In conclusion, our findings suggest that the composition and consistency of the desired affective state are likely dependent upon personality, motivation, and, potentially, an interaction between personality and motivation. Although further research is required to concretize exactly how these constructs shape the content of the desired affective state, they do to play a key role in determining how one wants to feel.

Acknowledgments

The authors would like to thank Jessica Bunn, Heather Grogan, the Axio survey system team at Kansas State University, and the AskWU survey system team at Washington University in St. Louis for their assistance with data collection.

Footnotes

1

A wealth of research does demonstrate that cultural differences exist in the DAS (i.e. Affect Valuation Theory: Tsai et al. 2006). However, cultural differences in the DAS are mediated to differences in culturally based motivation (Tsai et al. 2007). As motivation is the explanatory factor for these findings, we choose not to focus on culture per se, but rather the higher order predictor of motivation.

2

Although coping in response to stress and affect regulation are neither triggered by the same circumstances nor carried out in the same manner, stress coping is still applicable to affective experience. The increased use of effective stress coping strategies has been consistently linked to the experience of more positive and less negative affect (see Folkman and Moskowitz 2000).

3

Gender does predict regulatory behavior such that men use certain ineffective affect regulation strategies more frequently (Gross and John 2003) and may show a lesser ability to regulate affective states (Augustine and Hemenover 2009). Although demographics could not be collected as a part of Study 1, Study 2 did reveal several gender differences in the DAS. Women desire less negative affect (NA: β = −.24, SE = .08, p < .05; UU: β = −.32, SE = .10, p < .05; U: β = −.24, SE = .08, p < .05; AU: β = −.33, SE = .09, p < .05), less high activation affects (β = −.29, SE = .12, p < .05) and less low activation affects (β = −.30, SE = .13, p < .05; all other βs = −.02 to .02, ns).

Portions of this paper were presented at the 8th annual meeting of the Society for Personality and Social Psychology (January 2007) and 20th annual convention of the Association for Psychological Science (May 2008).

Contributor Information

Adam A Augustine, Department of Psychology, Washington University in St. Louis, Campus Box 1125, One Brookings Drive, St. Louis, MO 63130-4899, USA.

Scott H. Hemenover, Western Illinois University, Macomb, IL, USA

Randy J. Larsen, Washington University in St. Louis, St. Louis, MO, USA

Tirza E. Shulman, University of Wisconsin, Fond du Lac, WI, USA

References

  1. Augustine AA, Hemenover SH. On the relative effectiveness of affect regulation strategies: A meta-analysis. Cognition and Emotion. 2009;23:1181–1220. [Google Scholar]
  2. Bonanno GA. Emotion self-regulation. In: Bonanno GA, Mayne TJ, editors. Emotions: Current issues and future directions. Guilford; New York, NY: 2001. pp. 251–285. [Google Scholar]
  3. Bushman BJ. Does venting anger feed or extinguish the flame? Catharsis, rumination, distraction, anger, and aggressive responding. Personality and Social Psychology Bulletin. 2002;28:724–731. [Google Scholar]
  4. Connor-Smith JK, Flachsbart C. Relations between personality and coping: A meta-analysis. Journal of Personality and Social Psychology. 2007;93:1080–1107. doi: 10.1037/0022-3514.93.6.1080. [DOI] [PubMed] [Google Scholar]
  5. Costa PT, Jr., McCrae RR. Influence of extraversion and neuroticism on subjective well-being: Happy and unhappy people. Journal of Personality and Social Psychology. 1980;38:668–678. doi: 10.1037//0022-3514.38.4.668. [DOI] [PubMed] [Google Scholar]
  6. Costa PT, McCrae RR. Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Psychological Assessment Resources; Odessa, FL: 1992. [Google Scholar]
  7. Diener E, Diener C. Most people are happy. Psychological Science. 1996;7:181–185. [Google Scholar]
  8. Folkman S, Moskowitz JT. Positive affect and the other side of coping. American Psychologist. 2000;55:647–654. doi: 10.1037//0003-066x.55.6.647. [DOI] [PubMed] [Google Scholar]
  9. Geen RG, Quanty MB. The catharsis of aggression: An evaluation of a hypothesis. In: Berkowitz L, editor. Advances in experimental social psychology. Vol. 10. Academic Press; New York, NY: 1977. pp. 1–37. [Google Scholar]
  10. Gross JJ. The emerging field of emotion regulation: An integrative review. Review of General Psychology. 1998;2:271–299. [Google Scholar]
  11. Gross JJ, John OP. Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology. 2003;85:348–362. doi: 10.1037/0022-3514.85.2.348. [DOI] [PubMed] [Google Scholar]
  12. Hemenover SH, Augustine AA, Shulman TE, Tran TQ, Barlett C. Individual differences in the ability to repair negative affect. Emotion. 2008;8:468–478. doi: 10.1037/1528-3542.8.4.468. [DOI] [PubMed] [Google Scholar]
  13. John OP, Donahue EM, Kentle RL. The big five inventory–Versions 4a and 54 (technical report) Institute of Personality Assessment and Research; Berkeley, CA: 1991. [Google Scholar]
  14. Kampfe N, Mitte K. What you wish is what you get? The meaning of individual variability in desired affect and affective discrepancy. Journal of Research in Personality. 2009;43:409–418. [Google Scholar]
  15. Larsen RJ. Toward a science of mood regulation. Psychological Inquiry. 2000;11:129–141. [Google Scholar]
  16. Larsen RJ, Augustine AA. Basic personality dispositions: Extraversion/neuroticism, BAS/BIS, positive/negative affectivity, and approach/avoidance. In: Elliot AJ, editor. Handbook of approach and avoidance motivation. Lawrence Erlbaum Associate; Hillsdale, NJ: 2008. pp. 151–164. [Google Scholar]
  17. Larsen RJ, Diener E. A multitrait-multimethod examination of affect structure: Hedonic level and emotional intensity. Personality and Individual Differences. 1985;6:631–636. [Google Scholar]
  18. Larsen RJ, Diener E. Promises and problems with the circumplex model of emotion. Review of Personality and Social Psychology. 1992;13:25–59. [Google Scholar]
  19. Larsen RJ, Prizmic Z. Affect regulation. In: Baumeister R, Vohs K, editors. Handbook of self-regulation research. Guilford; New York: 2004. pp. 40–60. [Google Scholar]
  20. McCrae RR. Aesthetic chills as a universal marker of openness to experience. Motivation and Emotion. 2007;31:5–11. [Google Scholar]
  21. McCrae RR, Costa PT., Jr. Adding Liebe und Arbeit: The full five-factor model and well-being. Personality and Social Psychology Bulletin. 1991;17:227–232. [Google Scholar]
  22. Meier BP, Robinson MD, Wilkowski BM. Turning the other cheek: Agreeableness and the regulation of aggression-related primes. Psychological Science. 2006;17:136–142. doi: 10.1111/j.1467-9280.2006.01676.x. [DOI] [PubMed] [Google Scholar]
  23. Nelissen RMA, Dijker AJM, de Vries NK. Emotions and goals: Assessing relations between values and emotions. Cognition and Emotion. 2007;21:902–911. [Google Scholar]
  24. Raudenbush SW, Bryk AS. Hierarchical linear models: Applications and data analysis methods. 2nd ed. Sage; Thousand Oaks, CA: 2002. [Google Scholar]
  25. Russell JA. A circumplex model of affect. Journal of Personality and Social Psychology. 1980;39:1161–1178. [Google Scholar]
  26. Rusting CL, Larsen RJ. Moods as sources of stimulation: Relationships between personality and desired mood states. Personality and Individual Differences. 1995;18:321–329. [Google Scholar]
  27. Shulman TE, Augustine AA, Hemenover SH. Studies in affect regulation: Linking affective chronometry and repair ability. In: Clark AV, editor. Psychology of moods: New research. Nova; Hauppauge: 2006. pp. 117–142. [Google Scholar]
  28. Swann WB, Stein-Seroussi A, Giesler B. Why people self-verify. Journal of Personality and Social Psychology. 1992;62(3):392–401. doi: 10.1037//0022-3514.62.3.392. [DOI] [PubMed] [Google Scholar]
  29. Tamir M. Don’t worry be happy? Neuroticism, trait-consistent affect regulation, and performance. Journal of Personality and Social Psychology. 2005;89:449–461. doi: 10.1037/0022-3514.89.3.449. [DOI] [PubMed] [Google Scholar]
  30. Tamir M, Chiu C, Gross JJ. Business or pleasure? Utilitarian versus hedonic considerations in emotion regulation. Emotion. 2007;7:546–554. doi: 10.1037/1528-3542.7.3.546. [DOI] [PubMed] [Google Scholar]
  31. Tobin RM, Graziano WG, Vanman EJ, Tassinary LG. Personality, emotional experience, and efforts to control emotions. Journal of Personality and Social Psychology. 2000;79:656–669. [PubMed] [Google Scholar]
  32. Tsai JL, Knutson B, Fung HH. Cultural variation in affect valuation. Journal of Personality and Social Psychology. 2006;90:288–307. doi: 10.1037/0022-3514.90.2.288. [DOI] [PubMed] [Google Scholar]
  33. Tsai JL, Miao FF, Seppala E, Fung HH, Yeung DY. Influence and adjustment goals: Sources of cultural differences in ideal affect. Journal of Personality and Social Psychology. 2007;92:1102–1117. doi: 10.1037/0022-3514.92.6.1102. [DOI] [PubMed] [Google Scholar]
  34. Watson D, Clark LA. On traits and temperament: General and specific factors of emotional experience and their relation to the five-factor model. Journal of Personality. 1992;60:441–476. doi: 10.1111/j.1467-6494.1992.tb00980.x. [DOI] [PubMed] [Google Scholar]
  35. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology. 1988;54(6):1063–1070. doi: 10.1037//0022-3514.54.6.1063. [DOI] [PubMed] [Google Scholar]

RESOURCES