Research on social influence proves that an important motive of compliance is the need for a positive evaluation from others. However, there is little knowledge of how people responding to social influence in a resistant way are assessed. This study including 187 people concerns the evaluation of communion and agency of people (protagonists of the story) who, in a situation of unwanted social influence, reacted either with consent or presented one of three types of resistance: reactance, scepticism or inertia. The results showed that the evaluation of the protagonist’s agency was highest when s/he reacted to persuasion with reactance, and lowest for those who behaved compliantly. However, the assessment of the communion of the same behaviours was completely opposite. Such substantial asymmetries between the evaluation of communion and agency of the protagonists reacting in different ways to attempts to influence were noted for each of the types of behaviour except for inertia. In this one case, the evaluation of communion and agency of the protagonist turned out to be almost identical. The results are reflected in terms of self-presentation, politeness theory and the importance of norms in evoking submission to social influence.

resistance to social influence, agency, communion, social perception, reactance, inertia, scepticism, self-presentation

In a world with abundant influences and influential experts, there is an increasing number of situations in which people do not want to be influenced by others, to undertake actions they are coerced into, to agree with an opinion which is forced upon them, to take advice, to accept reasoning or even to consent to somebody’s request. Nevertheless, influence is rarely rejected in a direct way, which is documented by a series of experiments based on the Asch or Milgram paradigm, along with their contemporary replications (e.g., Burger, 2009; Doliński et al., 2017; Kundu & Cummins, 2013; Mermillod, Marchand, Lepage, Begue, & Dambrun, 2015; Mori & Arai, 2010) and by data concerning the level of efficiency of some techniques of influence (e.g., 90% obedience when the disrupt-then-reframe technique is combined with the foot-in-the-door technique – Fennis, Das, & Pruyn, 2004). The scale of obedience is considerably larger than that of resistance.

The reasons for these findings can be found in the social nature of human life, in our crucial needs: maintaining bonds with others, obtaining acceptance and positive thinking about oneself (Cialdini & Goldstein, 2004; Doliński, 2005). The costs of resistance to others can be corroborated by the results of numerous studies conducted by, for example, Cacioppo’s research team (Cacioppo & Hawkley, 2014; Cacioppo et al., 2013) concerning isolation, rejection and social exclusion and also by studies which analyse the consequences of whistle-blowing (Bjørkelo, 2013; Jackson et al., 2010). In less dramatic situations, resistance or even mere refusal infringe upon rules of politeness and require some sort of justification, which is why compliance can simply be easier (Bohns, 2016; Bohns & Flynn, 2010; Flynn & Lake, 2008). A high cost of resistance against influence finds confirmation in the research on brain structure activity by means of neuroimaging /fMRI/ (Berns et al., 2005).

According to self-determination theory, autonomy, independence and control over the situation constitute an exceptionally important aspect of the self (Deci & Ryan, 1985), being also personal traits that are highly valued in others. For example, external observers evaluated people resisting to perform a morally dubious task more positively in comparison to those who were obedient (Monin, Sawyer, & Marquez, 2008).

Studies undertaken in this field, which are considerably less numerous than those regarding the efficiency of techniques of influence, have focused on various areas in which resistance can be manifested and operationalised in various ways. Arguably, resistance to influence can be related both to characteristics evaluated in a positive way and to those that produce much worse associations. It was noted that obedience is manifested or hidden for strategic purposes, i.e., to make an impression on the environment or to protect a sense of identity (Baer, Hinkle, Smith, & Fenton, 1980; Lee, 2013; Nail & MacDonald, 2007). Resistance to influence can bring substantial losses. However, just like compliance it can also generate some gains in terms of image. Taking into consideration mechanisms of social perception, it is important by whom and in what dimensions an individual resisting influence is evaluated: image losses in one dimension may be compensated in another.

The present study investigated social perception of different types of resistance to social influence. On the basis of Knowles and Riner’s concept (2007) developing the previous theses by Knowles and Linn (2004a, 2004b), we examined the ratings of people who, in a situation of unwanted influence, comply or show resistance in the form of reactance, scepticism or inertia in the agency and communion dimension (Wojciszke, 2010).

The collected data was analysed using variance analysis with repeated measurements5. The role of the inter-group factor was played by the version of the story that illustrated reaction to influence (reactance, scepticism, inertia or consent). The repeated measurements are 2 aspects of the evaluation of the protagonist of the story (agency and communion). In this way, 8 conditions were distinguished (4 versions of the story × 2 aspects of evaluation), for which the descriptive statistics in the form of averages and standard deviations are presented in Table 1.

Table 1Download as CSV 

The Average Ratings of the Protagonist’s Agency and Communion Depending on Her/His Response to the Social Influence (Version)

	Agency		Communion
Version	M	SD	M	SD
consent	-0.55	0.46	0.85	0.53
inertia	0.26	0.60	0.37	0.37
reactance	0.60	0.65	-0.21	0.49
scepticism	0.47	0.57	0.16	0.60

The performed variance analysis showed insignificant main effects of the story version, F (3, 183) = 2.44, p = .066, η² = .038 and the repeated measurement, F (1, 183) = 2.30, p = .131, η² = .012. However, the interactional effect of the story version and repeated measurement is crucial for the verification of the hypotheses. The effect proved to be statistically significant, F (3, 183) = 68.02, p < .001, η² = .527, which is shown in Figure 1. All analyses assumed a 95% level of confidence.

Contrast analyses showed significant differences in the evaluation of the protagonist manifesting each of the resistance reactions vs. consenting reactions in the agency domain (higher agency for resistance reactions). A significantly lower evaluation of agency in the case of consent in comparison with a higher evaluation of this dimension for an inertial response, F (1, 183) = 44.96, p < .001; with a sceptical response, F (1, 184) = 73.31, p < .001, and the highest score of agency with reactance reaction, F (1, 183) = 91.11, p < .001, confirm hypothesis H1. A significantly lower evaluation of agency in the case of inertial behaviour in comparison with the highest evaluation of this aspect for reactant response, F (1, 183) = 8.34, p = .004, was also noted. However, there was no significant diversification between the agency of protagonists showing reactance and scepticism (an insignificantly higher agency was assigned to a reactant person), F (1, 183) = 0.81, p = .369, and between inertial and sceptical responses, F (1, 183) = 3.83, p = .052 (higher agency was assigned to a sceptical protagonist – the results close to statistical significance). In conclusion, the results are only partially compatible with H1a) hypothesis – with regard to the relatively lower evaluation of the agency in inertial protagonists compared to other reactions of resistance.

Figure 1.

The perceived agency and communion of the protagonist depending on his/her response to the social influence

Contrasts analyses also showed significant differences in the evaluation of the protagonist manifesting each of the resistance reactions vs. consenting reactions in the communion aspect (higher communion for consent). A significantly higher evaluation of communion in the case of consent in comparison with the lower evaluation of: inertial response, F (1, 183) = 20.80, p < .001, sceptical F (1, 184) = 41.42, p < .001, and reactant behaviour, F (1, 183) = 100.68, p < .001, confirm hypothesis H2. A significantly lower evaluation of communion in the case of reactance in comparison with each form of resistant responses was also noted; The assessment of communion of the person manifesting scepticism (M = 0.16, SD = 0.60) was significantly higher than in the case of reactance F (1, 183) = 12.61, p < .001, and also significantly lower than the evaluation of an inertial protagonist, F (1, 183) = 24.97, p < .001. Among the persons who reacted to the attempts of influence with resistance, the highest level of communion was attributed to those who behaved in an inertial way (M = 0.36, SD = 0.37). The results of contrast analyses show that hypothesis H2a) is confirmed in this study.

Finally, contrast analyses were applied to look for differences between the two aspects (agency and communion) of the evaluation for each version. It was found that only for inertia were there no significant differences, F (1, 183) = 0.85, p = .357, Significantly higher evaluations of agency than communion were found for reactance, F (1, 183) = 52.53, p < .001, and scepticism, F (1, 183) = 8.52, p = .004, while for consent – significantly higher evaluations of communion than agency, F (1, 183) = 143.67, p < .001, i.e., according to the assumptions of H3.

As shown above, the main effect of the version factor in the applied model of analysis proved to be insignificant since it takes into account the joint evaluation in terms of agency and communion. As seen in the chart, agency and communion constitute almost a mirror image of themselves so, if treated jointly, they mutually eliminate the diversification. Only the analysis of the interaction between the story version and an aspect of evaluation (agency/communion), supported by the analysis of contrasts, showed differences in the evaluation of protagonists reacting by way of reactance, scepticism, inertia or consent, with regard to every aspect of the evaluation. It may be concluded that the version of the story differentiates the evaluation of the protagonist, if every aspect of evaluation is considered individually. This can be proven by additionally performing two separate one-factor analyses, where the independent variable is the version of the story, while the dependent variable is only agency or only communion. In both cases the relevant main effects are obtained: for agency, F (3, 183) = 36.65, p < .001, η² = .375, while for communion, F (3, 183) = 34.91, p < .001, η² = .364, the description of which is unnecessary since it would repeat the conclusion of the analysis of interaction, which has already been discussed.

The obtained results confirmed the assumptions about different social perceptions of showing various reactions to social influence. The protagonist of the story, persuaded by an acquaintance to adopt a diet and a schedule of exercises, was evaluated in a different way depending on whether they consented unwillingly or showed resistance.

Reactant behaviors turned out to be perceived as the most agency related, skeptical -only slightly less so. The evaluation of inertial behaviour in the agency domain was significantly lower than that of the other two reactions of resistance, but was still significantly higher than the ratings obtained for consenting behaviour.

These results can be related to the perception of activeness on the part of a subject of influence and directness of the manifestation of his unwillingness or objection to persuasion, which is compatible with, for example, Coetsee’s approach who, specifically in the dimension of activeness, describes various reactions to the introduction of changes in an organisation (Coetsee, 1999). The assessment of agency is dominated not only by the perceived activeness of the subject of influence but also by the factor of their control over the situation; the high position of reactance (active objection) and scepticism (an attempt to persuade the agent of influence that the offer has some weaknesses) lend support to the argument. Moreover, control over a situation is strictly connected to a relation of power, and powerholders are usually highly evaluated with regard to agency/competences (Anderson, Hildreth, & Howland, 2015). Resistance in the form of inertia concludes, in a sense, the episode of influence and terminates its exertion; now the influenced person regains control over the situation, which is not the case for consent, as described in the research.

Also Koestner et al. (1999) showed that people who are inclined toward reactive autonomy interpret attempts to persuade them as a threat to their self-determination. Also it seems probable that observers of such interactions attach importance to the degree to which the person being watched holds control over the situation.

In the realm of communion, the consenting subject received the highest rating, while reactant behaviour scored the least points. The top score of consent is almost self-explanatory. An interesting case is presented by inertial reaction to unwanted influence, which appears to be communion-related. Arguably, the lack of a directly expressed objection enables the perception of the actor as being respectful of the rules of politeness towards others and being considerate of their well-being. According to politeness theory (Brown & Levinson, 1987) inertial reaction allows the agent (persuader) to save face as no personal conviction is being expressed by the subject. In turn, scepticism could infringe on the rules of politeness since doubts expressed about a person’s proposition suggest unreliability or the incompetence of the proposer and may lead to shaming him. Since it is only an indirectly expressed suggestion, the evaluation of a sceptical reaction in the communion dimension yields a significantly higher score than an outright objection, which is characteristic of reactance. It seems that the lowest score of a reactance protagonist in the realm of communion results from the contravention of the norm of reciprocity - the force of which was frequently emphasised in establishing compliance (Cialdini & Goldstein, 2004; Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007).

The results obtained in our research indicate a lack of significant differences between the evaluation of skeptical and inertial reactions to social influence (in terms of both communion and agency). Admittedly, Knowles and Riner (2007) make a marked distinction between them; however, they refer mainly to the motivation lying at their root. Apparently, this distinction is of little importance in terms of social perception.

The study has shown a marked asymmetry between the perception of agency and communion with regard to particular types of resistance to influence and consent to undertake actions compatible with the persuader’s intentions. The more a given behaviour was perceived as communion-related, the lower score it obtained in the agency dimension. This is demonstrated in the present study, and especially in the scores for the reaction of consent to social influence. Its low rating in the agency dimension could have many reasons.

In the story presented to respondents the status of its protagonists was similar and the points of emphasis were placed on reluctance on the part of the subject of influence (thinks that it is a very bad idea). Consent can be interpreted in this regard as unnecessary submissiveness since there are no grounds to think that it is employed to obtain some benefit. Therefore consent can be perceived as weakness, irrationality, a mistake or lack of control over the situation. Still there are some benefits, if only in terms of a comfortable social interaction and a good rapport. Although this benefit is reaped by the agent of influence, the norm of reciprocity, which is fundamental to the existence of communion, is satisfied.

A strong asymmetry also concerns reactance, which is highly rated as being agency-related but scores very poorly in relation to communion. A possible explanation with respect to communion was discussed earlier; the explanation of its high score for agency, apart from such factors as activeness and power, could be that people who behave in this way focus on their own needs and interests which they actively protect. Possibly, some inconsistencies obtained in the results of research on self-presentational aspects of reactance, as discussed by, inter alias, Nail, Van Leeuwen, and Powell (1996), can be attributed to the lack of consideration for the variation in the obtained effects between the two dimensions of social perception.

There is a strong effect of negativity in evaluating the morality of other people, i.e., negative items of information are considered more diagnostic, more important and sought after in comparison to positive ones. The effect is not found, however, with reference to competence-related traits (Czapiński, 1988; Wojciszke, 2002) where an opposite effect – that of positivity - is mentioned (Wojciszke, 2002). Consequently, when striving to be positively perceived the best course of action is to hide personal moral vices that might harm the communion and to promote intellectual values. Inertia and scepticism, among all the strategies of resistance, come closest to these recommendations. Inertia gives protection against social condemnation because it stops short of explicit refusal. Although inertia may look like avoidance, it helps protect individual interest, without infringing upon the interests of others. The present study indicates that it is a relatively low-risk strategy. Among all the forms of resistance, inertial behaviour produces the smallest loss in one’s communion-related rating. Another advantage of inertia is the possibility of stopping further attempts to exert influence upon an individual, thus exercising non-invasive control over the situation. The advantage of scepticism is, in turn, its intellectual, rational nature, which may result in the evaluation of a sceptical person as a clever person (cf. Czerniak, 2009). Even if the intentions of a person reacting sceptically are, to some extent, directed against the sender it cannot be ruled out that they serve the common purpose of preserving collective interest and that the benefits may not only be obtained by the sceptically reacting person but also by other people (i.e., other possible addressees of the request directed at the sceptic which, as a consequence of the argumentation, will have to be made more sensible or simply withdrawn).

The least profitable strategy in terms of image in the communion dimension is reactance, which received the highest score in the agency domain. Resistance, as mentioned above, is not a frequently expressed reaction, except in teenage rebellion (Oleszkowicz, 2006), as it exposes an individual to social disapproval and rejection and is related to numerous costs. Scepticism, in turn, entails costs related to the intellectual effort necessary to generate content-related arguments for the rejection of the offer made by the agent of influence. In social terms, a sceptic bears the risk of becoming perceived as a person who splits hairs, plays the wise guy, etc. Inertial reaction can bring costs related to being insincere towards the person trying to exert social influence; however, it was proven that it does not entail unfavourable social reception in terms of competence or morality (it is surprisingly equivalent in both domains); however, it does not contribute to achieving a very high rating.

As demonstrated in the present study, the discrepancies between the evaluation of reactions in the domains of communion and agency, as well as the fact that social costs related to a lower rating in one are compensated by gains in the other – are compatible with the previous findings about the relationship between the two domains. Peeters (2008), using the terms other- vs. self-profitability, raises the issue of extremity/moderation of traits. Apart from the fact that extremities tend to be rated negatively, while moderate traits – positively, when moving along this continuum, there is a curious shift between the dimensions: for example, a negative trait pertaining to an individual benefit: extravagant or passive, when mitigated, becomes a positive trait regarding the benefit of others: generous or peaceful. A similar negative trait pertaining to the benefit of others, i.e., stingy or aggressive, when mitigated, becomes a positive one regarding an individual benefit: thrifty or forceful. Conversely, positive traits of moderate intensity, in their extreme manifestation, become negative, but in a complementary dimension. Thus between the extremes in one dimension there is a sort of intermediary link with a complementary dimension.

Our results seem to be compliant with the above statements: between the rejection of influence (disadvantageous to communion) and submission to it (disadvantageous to agency), there seems to be a place for inertial “basically yes, but not just yet.” Also the specialist literature on the subject of social influence (Nail et al., 2013) indicates that in the studies performed so far there has been no attempt to consider compromise as an indirect reaction between various forms of acceptance and rejection of influence, such as independent behaviours, anti- or non-conformist or reactant ones. In this perspective, inertial reaction to influence seems to perform such an indirect function of compromise. The results of research discussed thus far in this paper warrant the following answer to the question brought forward in the title: inertia is identified as a type of resistance against social influence that is least harmful to the image of those reluctant to yield to influence.

Given the paucity of sources published on the subject, our research performs an exploratory function. The stories presented to the respondents described a relatively “weak” situation of influence: the influence was direct but unconnected with a relation of power, was exerted by an acquaintance - a person familiar enough to visit the subject of influence; however, the protagonists were not related by strong interpersonal bonds and the influence concerned an issue of little importance to the influenced party (the request was neither pro- nor anti-social), although potentially positive in nature (the implementation of the program of exercises and a diet). We assumed that if differences emerged in the evaluation of people who consent and who oppose in a ‘weak’ scenario of influence like ours, in cases of stronger influence we would obtain differences of degree rather than of quality. Nevertheless, the area of study requires further research that might develop along the following lines: the use of requests and instructions of a pro- and anti-social nature (to manipulate the force of coercion thus presenting a varying degree of justification to resistance); adoption of the influenced person’s perspective (unlike in the present study – that of an observer) and focusing research on inclinations to choose a specific strategy of resistance or consent in response to factor manipulation. Another worthwhile research task would be to diversify the descriptions of the stories with a view to bringing out the respective interests of the sender and the receiver of influence.

An important limitation of our study is the fact that only fictionally described protagonists were assessed instead of people who act in real life. Accordingly, in the next investigations it would be worth focusing on the actual behaviour of people who resist influence and on those witnessing such behavior.

In addition to outlining the direction of further research, the findings can also have a practical application: knowing that the assessment of people with different responses to social influence is compensatory in nature - lowering it in one dimension entails raising it in another one - can help, for example, in solving self-presentation dilemmas, in encouraging people to resist negative influences, and can be useful in projecting a more effective impact on others in the areas of positive social importance (health, education, etc.).

The research has been financed from the resources for the statutory activity of the Faculty of Psychology and Humanities WPiNH/DS/5/2016, continued in the year 2017 (WPiNH/DS/5/2016-KON).

The authors have declared that no competing interests exist.

The authors have no support to report.

The data analyzed in this paper are freely available via the PsychArchives repository. For further information see the "Supplementary Materials" section.

Data and codebook of this study are accessible via the PsychArchives repository.