Two-Factor Model of Personality
The two-factor model (TFM) of personality is a model of personality traits structure that was discovered through factor analysis of traits, with two broad factors (also known as metatraits) emerging at the highest level (Digman 1997; Saucier et al. 2014). These factors/metatraits have been found to possess a theoretical meaning that corresponds to many psychological constructs developed and used in personality research (Digman 1997) and to have the potential to integrate the various concepts (Saucier et al. 2014; Strus and Cieciuch 2017). The TFM was developed in the course of research on the Big Five traits (Goldberg 1990; McCrae and Costa 2003) and can be treated either: (1) as its continuation and extension (DeYoung 2015) or (2) as a step beyond this tradition, opening new possibilities for personality structure description (Strus and Cieciuch 2017).
Dual constructs in psychology: Inspirations, examples from various fields of psychology, and two-factor models in the psychology of personality traits
Inspirations for conceptualizations and labels
Agency and communion as basic modalities of human existence
Plasticity and stability as properties of learning systems
Example dual constructs in different fields of contemporary psychology
Basic dimensions of social cognition
Agency and communion
Abele and Wojciszke 2014
Power and intimacy
Basic human values
Openness (vs. conservation) and self-transcendence (vs. self-enhancement)
Schwartz et al. 2012
Basic dimensions of interpersonal interaction
Agency/status and communion/love
Wiggins and Trapnell 1996
Basic properties of ego
Ego-resiliency and ego-control
Block and Block 1980
Basic development processes
Accommodation and assimilation
Basic dimensions of affect
Positive and negative affect
Watson and Tellegen 1985
Basic dimension of temperament
Impulsivity (BAS) and anxiety (BIS)
Basic types of psychopathology
Internalizing and externalizing problems
Krueger and Markon 2006
Basic types of self-perception biases
Superhero (egoistic bias) and saint (moralistic bias)
Paulhus and John 1998
Two-factor models: Basic dimensions of personality
Beta and alpha
Plasticity and stability
DeYoung et al. 2002
Dynamism and virtue
De Raad and Barelds 2008
Dynamism and self-regulation
Saucier et al. 2014
Outside the Big Five paradigm
Mental health and behavior control
In the psychology of personality and individual differences attempts have been made to combine both approaches. One of the most important efforts in this area was the five factor model of personality (FFM; McCrae and Costa 2003), which was constructed as a theoretical interpretation of the Big Five discovered within psycholexical studies. Thus, the FFM utilized results from inductive psycholexical research, while taking into consideration theoretical studies and inputs from other concepts and models of personality (McCrae and Costa 2003). For some time, the FFM was thought to describe universal, orthogonal dimensions of human personality and it was expected that it could serve as a unifying reference framework for other models of traits and various constructs which could be located within its five-dimensional space. However, we now know that the FFM is not entirely free from shortcomings, either. At the conceptual level, it has been criticized for its atheoretical nature (e.g., Block 2001), and at the empirical level for the systematic nonorthogonality of the proposed personality dimensions.
Currently, efforts are being made to combine the inductive and deductive approaches under the TFM, which not only exhibits the strengths of both approaches but also overcomes the shortcomings of the FFM. The TFM describes the personality traits as organized within two broad factors revealed through the factor analysis. There are different versions of the model varying, inter alia, in (1) the variables included in analysis, (2) the theoretical meaning ascribed to them, and (3) the role the two factors play in a comprehensive description of personality. Nevertheless, all the variants share the common idea that many personality traits are organized at the highest level in two factors, which are very general dimensions of personality.
The TFM stems from an inductive quest for completeness and comprehensiveness of personality description, which has been implemented through psycholexical research; at the same time it incorporates many theoretical intuitions from the deductive approach, which gave rise to many dual constructs (see Table 1). This convergence seems rather surprising as it was not actively pursued by either paradigm. Nevertheless, today it can be considered a milestone in the personality psychology as it revealed two fundamental dimensions of personality discovered in research on its trait structure (including both psycholexical studies leading to the Big Five and questionnaire surveys resulting in the FFM) and interpreted in terms of many previously developed theoretical models which proposed pairs of variously termed constructs. In the TFM, those dimensions are known as alpha and beta (Digman 1997), stability and plasticity (DeYoung et al. 2002), or social self-regulation and dynamism (Saucier et al. 2014).
Theoretical Context: Dual Constructs in Psychology
Various models distinguishing two basic constructs have been developed in many fields of personality psychology. Similarity between those models has become much clearer and more obvious since the discovery of the TFM in research on the structure of personality traits, as it has enabled the incorporation of many theoretical intuitions included in the dual constructs. In the current literature, researchers refer to two sources outside psychology that have inspired thinking in categories of two dimensions, namely philosophy and computer science. The first one involves Bakan’s (1966) book under the telling title The Duality of Human Existence, which argues that there are two fundamental modalities not only in humans but in all forms of life: agency for the existence of an organism as an individual, and communion for the participation of the individual in some larger organism of which the individual is a part (p. 15). The second source involves the stability-plasticity dilemma posed by Grossberg (1980), which describes the situation of artificial and biological learning systems tasked with learning new things without forgetting the old ones. Plasticity gives the ability to acquire new knowledge, but too much of it may result in catastrophic forgetting, causing the loss of previous memories (Grossberg 1980). Thus, stability prevents the learning system from catastrophic forgetting. These two pairs of constructs: agency-communion and stability-plasticity seem to denote quite different phenomena, but both of them continue to inspire research on the TFM.
The humanistic constructs proposed by Bakan (1966) refer to human nature, while Grossberg’s (1980) cybernetic constructs primarily concern cognitive systems, although they can also be extended to other entities. DeYoung et al. (2002) popularized the two cybernetic terms in personality psychology by applying them to the two factors discovered by Digman (1997), which were in turn named metatraits (DeYoung 2015). Although Bakan’s (1966) terms were also initially used in personality psychology (e.g., Wiggins and Trapnell 1996), in the current literature they are mostly found in social (and especially sociocognitive) psychology, for instance in the model proposed by Abele and Wojciszke (2014), where agency and communion refer to the content of social perceptions (the categories proposed by Bakan are close to what people see, think, and experience). Again, contrary to the expectation that cybernetic terms would be more appropriate for sociocognitive psychology, they appear to be more useful in research on the structure of personality, where the metatraits stability and plasticity are used to describe how personality operates at the most general level (DeYoung et al. 2002; DeYoung 2015).
Along with Bakan’s (1966) and Grossberg’s (1980) terms, Table 1 presents examples of dual constructs from different areas of psychology (for more on dual constructs see: Digman 1997; Paulhus and John 1998; Wiggins and Trapnell 1996).
Most of these dual constructs were proposed as theoretical models of various phenomena. They were usually developed in the deductive tradition, with theoretical hypotheses being empirically validated. These constructs have been (or can be) used to interpret the theoretical content of the two factors that were discovered (and confirmed) both in psycholexical and questionnaire studies on the structure of personality traits by means of factor analysis (cf. Digman 1997; Saucier et al. 2014).
Empirical Discovery: The Big Two as Basic Dimensions of Personality
Some of the basic questions in personality research are how many traits constitute fundamental dimensions of personality and what their structure is. Studies addressing these issues have been conducted within two frameworks, namely, the psycholexical and questionnaire paradigms, and for some time they culminated in a consensus that there exist five most general, fundamental, and orthogonal dimensions, known as the FFM in the questionnaire paradigm and as the Big Five in the lexical paradigm. The latter term is also often used as a shorthand for the five dimensions irrespective of their theoretical underpinnings.
However, the existence of five orthogonal dimensions has recently been questioned by the TFM, which postulates two broader factors above the Big Five. Indeed, after being confirmed in the lexical and questionnaire research paradigms, the Big Five was subsequently challenged within both frameworks in favor of the TFM. In what follows, we briefly present the two research approaches, which first discovered the Big Five, and then, as a result of continued studies, gave rise to the TFM.
The Big Five
The Big Five was discovered in psycholexical research founded on the lexical hypothesis, according to which all the important personality traits are reflected in language (Goldberg 1990). Thus, one may investigate the linguistic traces of personality and determine their structure. This can be done by means of factor analysis, reducing a large number of variables to a few dimensions. In this case, the multiplicity of linguistic expressions for personality traits was reduced to five basic dimensions (Goldberg 1990). Due to the inductive atheoretical approach used, the factors were simply labeled with Roman numerals (I through V), which is consistent with inductionism, an approach in the philosophy of science according to which one should avoid any theoretical assumptions to come as close as possible to the facts. The observations gathered in this way are then generalized and turned into universal statements. In psychology, this methodological paradigm seems to be best exemplified by psycholexical research, as studies on the structure of personality began with analysis of facts (words describing personality) followed by efforts to establish the structure of the traits conveyed by those words.
The meaning of the identified factors was determined on the basis of adjectives with the strongest loadings in factor analysis, and their order followed from the number of adjectives that each of them comprised. In other words, Factor I had the largest linguistic representation, while Factor V was the least frequently represented. Factor I is usually called surgency or extraversion, Factor II – agreeableness, Factor III – conscientiousness, Factor IV – emotional stability, and Factor V – culture, intellect, or imagination (Goldberg 1990).
Efforts to identify the structure of personality traits were undertaken not only within the psycholexical framework, but also under the questionnaire paradigm, which sought to develop the optimum model of personality structure and dimensions pursuant to some theoretical assumptions. This led to, inter alia, the Giant Three model, which was termed PEN (psychoticism-extraversion-neuroticism) by Eysenck (Eysenck and Eysenck 1985) and interpersonal circumplex by Wiggins (see Wiggins and Trapnell 1996). Observing that most of the existing psychometric personality questionnaires (including the instrument measuring the traits identified by Eysenck and Eysenck 1985) contained a measure of neuroticism (the opposite of emotional stability) and extraversion, Costa and McCrae adopted them as the fundamental dimensions of personality (McCrae and Costa 2003). During further research, they added the dimension of openness to experience (thus forming the NEO model). Subsequently, inspired by the results of psycholexical studies, they also included agreeableness and conscientiousness. Finally, they constructed scales for measuring all the dimensions (the NEO-PI), which were later revised as the NEO-PI-R (McCrae and Costa 2003).
In this way, the psycholexical model of the Big Five was incorporated in questionnaire research. The FFM, consisting of neuroticism (vs. emotional stability), extraversion, openness to experience (counterpart of Intellect), agreeableness, and conscientiousness, was conceptualized in terms of lower-order traits (facets) and operationalized in questionnaires. Within this approach, research on personality structure was expanded both theoretically and empirically. Since then, the five factors have been treated not only as merely descriptive personality traits (Saucier and Goldberg 2001), but they have also been interpreted in terms of dispositions or tendencies to certain patterns of thoughts, emotions, and behaviors, and the relationship of the FFM to other models and theoretical concepts has been studied (McCrae and Costa 2003). Finally, McCrae and Costa (2003) developed the Five Factor Theory (FFT) of personality on the basis of the FFM.
The FFT treats the FFM as biologically based basic tendencies, which constitute one of two core components of a personality system and influence characteristic adaptations (the second core component, comprising the self-concept). The latter are in turn related to other interfacing elements (objective biography and external influences). These one-way or mutual influences or relationships form causal pathways in which dynamic processes operate.
Discovery of the Big Two in Psycholexical Research
The Big Five model was discovered in psycholexical studies conducted in the English language (Goldberg 1990), and the replication of the Big Five has encountered substantial problems in non-Germanic languages despite continued efforts since the 1990s (e.g., De Raad et al. 2010). Some of the findings pointed to the existence of a sixth factor, giving rise to the Big Six model, also known as HEXACO in the questionnaire approach (Ashton and Lee 2007; Saucier 2008), while others indicated that replicability was possible only for a smaller number of basic factors (cf. De Raad et al. 2010; Saucier 2008; Saucier and Goldberg 2001). Large-scale psycholexical studies based on systematic taxonomic comparisons of more than a dozen different languages have shown that only the Big Three factors were replicable (De Raad et al. 2010).
Indeed, there is now a considerable body of evidence that only two broad factors appear to be fully ubiquitous across languages and cultures (Saucier 2008; Saucier et al. 2014). The Big Two factors were found in data from the lexicons of several languages belonging to different linguistic families or groups, such as Chinese, Korean, Filipino, Turkish, Greek, Polish, Hungarian, Maasai, and Senoufo (Saucier et al. 2014), representing diverse cultural characteristics and spoken on different continents, including Europe, North America, Asia, Africa, and Australia (Saucier et al. 2014; cf. Saucier 2008; Saucier and Goldberg 2001). These factors can be replicated despite differences in the criteria and procedures of word selection, factor extraction, replicability assessment, research method, and data collection (Saucier 2008; Saucier et al. 2014). It should be stressed that the psycholexical Big Two were not derived as higher-order factors from the Big Five; instead, they emerged in exploratory factor analysis which extracted two factors at the first-order level from hundreds of natural language trait descriptors (cf. Saucier 2008; Saucier et al. 2014). The Big Two may be further subdivided into three, five, or more traits.
One of the cross-culturally universal dimensions of the Big Two has been variously labeled virtue (De Raad and Barelds 2008), morality or social propriety (Saucier 2008; Saucier and Goldberg 2001), and recently social self-regulation (or S dimension; Saucier et al. 2014), while the other factor has been invariably termed dynamism (or D dimension). According to the study of Saucier et al. (2014), the key adjectival markers of the Big Two factors in the lexicons of nine diverse European, Asian, and African languages (see above), were: good, obedient, kind, generous, respectful, honest, selfish (−) [(−) denotes inverse pole], diligent, gentle, and responsible in the case of the social self-regulation factor, and weak (−), bold, brave, shy (−), timid (−), active, and lively in the case of dynamism. These 17 adjectives were the most consensual, i.e., they were found to be salient, recurrent markers in five of the nine languages. With slightly more relaxed criteria (presence in four of the nine languages), the additional markers were: careful, calm, patient, polite, and disciplined (for S dimension), and strong, fearful (−), sad (−), silent (−), energetic, cowardly (−), daring, sociable, cheerful, pessimistic (−), and dynamic (D dimension). However, while this set of markers captures the core lexical content of the Big Two, it does not cover the full breadth of their meaning (Saucier et al. 2014).
In general, social self-regulation includes attributes associated with socialization, solidarity, communion, cohesion, and adhering to sociomoral rules, and so it concerns morality, social propriety, and respect for others and for authority. It is also linked to using social norms as standards for regulating one’s own behavior. Dynamism includes attributes valuing active qualities and individual ascendancy, pertaining to liveliness, self-expression, self-confidence, and skills useful in dealing with social situations. It is associated with the activation-inhibition ratio and reward-punishment sensitivity. Therefore, while D dimension shows affinity to biological and process variables, revealing its temperamental core (e.g., approach-avoidance system), S dimension exhibits socialization and moral-ethical meaning, and so it involves the internalization of social and cultural norms, being predominantly a character-related dimension (Saucier et al. 2014; cf. De Raad and Barelds 2008; De Raad et al. 2010).
The psycholexical TFM can be treated as a parsimonious model of the basic bivariate structure of personality attributes (Saucier et al. 2014). The factors S and D are broad dimensions incorporating the overlapping content of various two-dimensional models or dual constructs, e.g., communion and agency, morality and competence, or externalizing and internalizing tendencies (see Table 1). For this reason, Saucier et al. (2014) argue that the psycholexical Big Two is a common-denominator model including the two most crucial axes of personality variation serving as a framework for linking diverse theoretical models and associated empirical findings.
Discovery of Two Higher-Order Factors in Questionnaire Research
Within the questionnaire tradition, the FFM was built on the psycholexical discovery of the Big Five; however, further questionnaire research on personality traits was conducted quite independently from psycholexical studies. From the very beginning, proponents of the FFM have had to defend the thesis that there exist exactly five basic dimensions of personality. Some researchers have proposed to change the number of dimensions, with Eysenck’s Giant Three (Eysenck and Eysenck 1985) and Ashton and Lee’s (2007) six factor HEXACO being the most serious alternatives in the 1990s and 2000s, respectively. Another challenge was to prove the orthogonality of the five factors, which ultimately led to the emergence of the TFM.
The orthogonality assumption was not confirmed, as the five factors were found to be correlated with one another (at levels often exceeding .40, or even .50), regardless of the measurement instrument used (e.g., McCrae and Costa 2003; Digman 1997). Digman (1997) was the first to notice there a pattern suggesting the existence of two higher-order factors. Interestingly, Digman (1997) discovered the TFM within the FFM paradigm using both exploratory factor analysis (typical of psycholexical research) and confirmatory factor analysis (usually applied in validating predefined models).
One higher-order factor was found to be related to the shared variance of emotional stability (vs. neuroticism), conscientiousness, and agreeableness, with the other one being responsible for the covariance of extraversion and openness to experience (intellect). This finding was based on reanalysis of more than a dozen studies conducted using different measures and informant types (teachers ratings, peer ratings, and self-reports) on samples originating from several countries and consisting of children and adolescents as well as young and mature adults. Digman (1997) named these factors alpha and beta, respectively, interpreting the former as a socialization factor and the latter as a personal growth factor in correspondence with the dual construct developed within the traditional and contemporary theories of personality (see Table 1).
Several years later the two higher-order factors of the Big Five were replicated and reconceptualized by DeYoung et al. (2002), who defined them as “metatraits” and relabeled them using the terms proposed by Grossberg (1980). Thus, alpha became the metatrait of stability, being linked to stability in the emotional domain (low neuroticism), motivational domain (high conscientiousness), and social domain (high agreeableness). In turn, beta was labeled the metatrait of plasticity, being meant to reflect behavioral (high extraversion) and cognitive (high openness) plasticity, reflected in the tendency to explore and voluntarily engage (behaviorally and cognitively) in new experiences (DeYoung et al. 2002). In other words, while the first metatrait is responsible for maintaining stability of psychosocial functioning, the second one involves exploration and adaptation to novelty and change. As such, the two higher-order factors (metatraits) of the Big Five have been supported in many studies (e.g., Chang et al. 2012; DeYoung 2006; cf. Strus et al. 2014).
Two Basic Personality Factors Identified Outside the Big Five Paradigm
The discovery of and research on the TFM took place mainly within the psycholexical and questionnaire traditions of the Big Five research. However, two broad personality factors were also found or confirmed in other studies and analyses carried out independently and outside the research lines discussed above. Of special importance is the series of studies by Becker (1999), who identified two higher-order factors of personality as mental health and behavior control, both being multifaceted constructs. Mental health, defined as the ability to cope with external and internal demands, encompasses: freedom from distress, meaningfulness, self-obliviousness, self-esteem, autonomy, capacity to love, flexibility and tenacity versus low physical and mental well-being, low coping ability, negative emotions, pessimism, self-centeredness, distrust, dependency, neuroticism. Behavior control is characterized by self-control (norm-, future-, and work-orientation, rationality, prudence, dutifulness, dependability, orderliness, frugality, conservatism and dogmatism) versus spontaneity (hedonism, excitement seeking, risk-taking, feeling orientation, changeability, radicalism, exhibitionism, dominance, egoism, and openness to experience). These two broad factors, also termed by Becker (1999) the Big Two, were found by extracting the first two orthogonal first-order factors from many analyses and reanalyses of large and representative pools of personality variables, as well as by extracting five or six oblique first-order factors and running a second-order factor analysis.
Mental health and behavior control do not seem to correspond well to the two broad personality factors described in the preceding sections. However, the results obtained by Becker (1999) led him to the conclusion that personality traits are organized in a circumplex structure, as many personality variables exhibit cross-loading on the Big Two higher-order factors. On this basis Becker (1999) has developed the circumplex model of higher-order personality factors. Treating Mental Health and Behavior Control as orthogonal axes of the circumplex, Becker (1999) proposed two other orthogonal, multifaceted dimensions: social adaptation versus unrestraint and self-actualization versus inhibition, located at 45° rotation to the former. The meaning and facets of these two additional higher-order factors correspond strongly to the two factors of personality discussed above, and especially to their versions developed by Digman (1997) and by the psycholexical approach (Saucier et al. 2014). Finally, Becker’s (1999) circumplex is an eight-octant model, with the octants being poles of four higher-order dimensions, according to which eight personality types can be defined (e.g., the high behavior control type, the low mental health type, the socially adapted type, and the inhibited type). Importantly, these types are clinically relevant, and the circumplex model itself is useful for the description of personality disorders, with valuable etiological hypotheses and therapeutic implications. In general, this line of research is continued in the circumplex model proposed by Strus and colleagues (Strus et al. 2014; Strus and Cieciuch 2017), although their model, in contrast to Becker’s, is directly based on the Big Five.
The Big One Above the Big Two?
Systematic intercorrelations between the five basic personality traits provided a point of departure for structural analysis which finally led to the discovery of the TFM. Analogously, the question arose as to whether the two basic dimensions are orthogonal or again systematically interrelated. If the latter were the case, this could suggest the existence of one general factor above the two previously identified ones. Indeed, Musek (2007) proposed the general factor of personality (GFP), analogous to factor g in the structure of cognitive abilities (intelligence). The GFP (Musek 2007; Rushton and Irwing 2011) was found to occupy the apex of a hierarchically organized structure of personality traits, and understood as a fundamental disposition integrating the most general, noncognitive dimensions of personality. The positive pole of the GFP was related to the optimal configuration of all functional aspects of personality, while its negative pole included a general tendency towards personality disorders and other psychopathologies (Rushton and Irwing 2011). Accordingly, the GFP could be characterized by high (vs. low) alpha/stability and beta/plasticity (Musek 2007; Rushton and Irwing 2011), and on the level of the Big Five by a configuration of high emotional stability, extraversion, openness to experience, agreeableness, and conscientiousness, although some findings have indicated that the GFP occupies the highest position in the hierarchical structure of personality regardless of the model adopted (Rushton and Irwing 2011).
The GFP exhibits an interesting theoretical meaning as it appears to be a general factor of mental health (Musek 2007; Rushton and Irwing 2011; cf. Becker 1999), revealing a strong relationship with social desirability, satisfaction with life, well-being, and self-evaluation (Musek 2007). Indeed, the GFP is claimed to be biologically based, whether in temperamental, neuropsychological, genetic, or evolutionary terms (Musek 2007; Rushton and Irwing 2011). According to the latter interpretation, the GFP is the effect of natural selection of socially desirable qualities, which facilitate functioning in a wide range of life situations, and which result from the K-strategy of reproduction in the context of Darwin’s theory of evolution and Wilson’s sociobiology (Rushton and Irwing 2011). Moreover, the GFP has a counterpart in the psycholexical approach termed the Big One, which is usually yielded with high intercultural replicability as the first factor in an unrotated structure (Saucier 2008). The psycholexical Big One includes a broad range of socially desirable traits (De Raad and Barelds 2008; Saucier 2008), and therefore has been labeled evaluation (Saucier and Goldberg 2001), socially desirable qualities (Saucier 2008), or virtue (De Raad and Barelds 2008).
However, in contrast to the TFM, which enjoys a rather widespread consensus in the literature (cf. Chang et al. 2012; DeYoung 2015; Musek 2007; Strus et al. 2014), the GFP hypothesis has been met with considerable criticism backed by studies questioning its existence (cf. Chang et al. 2012; Revelle and Wilt 2013), according to which it is an artifact or a method factor resulting from a social desirability bias.
In this context, of crucial importance are those studies on personality structure which apply multi-trait, multi-informant, and multi-instrument procedures. The use of more than one informant (self-reports combined with peer ratings) and more than one measure of the Big Five traits has not only corroborated the existence of alpha/stability and beta/plasticity, but has also proved any correlations between them to be artifacts, as the two factors are not correlated in the latent space of multi-informant models (DeYoung 2006; see also Chang et al. 2012). Empirical evidence for the orthogonality of these two basic factors undermines the existence of one general factor located at the top of a hierarchically organized structure of personality.
Toward Further Elaboration of Theoretical Meaning
Theoretical Meaning and Usefulness of the Two Factors
The Two-Factor model has been found to be a significant contribution to the description, understanding, and explanation of personality, as the two basic factors exhibit some unique psychological meaning, irreducible to the meaning of combinations of the Big Five traits (Strus and Cieciuch 2017). These two factors, or metatraits, refer to some very general patterns of behavior and experience (DeYoung 2006), and empirical research has shown that this level of description may be appropriate for the analysis, understanding, and prediction of other broad constructs (see DeYoung 2015; Saucier et al. 2014; Strus and Cieciuch 2017).
Biological foundations – alpha/stability and beta/plasticity have been found to be genetically determined and are thought to have neurobiological substrates in the serotonergic and dopaminergic systems, respectively (DeYoung 2006; DeYoung et al. 2002).
Theoretical explanatory mechanisms (DeYoung 2006, 2015; DeYoung et al. 2002) – alpha/stability and beta/plasticity are considered to describe, e.g., the broadest psychological properties (parameters) of the human cybernetic system (DeYoung 2015).
Possibility of integration with many other constructs developed within various models and theories of personality, emotion, and motivation (Digman 1997; Saucier et al. 2014; Strus and Cieciuch 2017) – see relationships of both alpha/stability and beta/plasticity, as well as social self-regulation and dynamism, with the constructs listed in Table 1.
In fact, the third point concerns the crucial mission of personality psychology, which is comprehensive integration or synthesis of knowledge generated within the some sub-disciplines of psychology. The constructs depicted in Table 1 relate to many different domains of human psychological functioning, e.g., basic properties of the ego, dimensions of motivations, values and affect, dimensions of social cognition and interpersonal interaction, as well as basic developmental processes and classes of psychopathology. Importantly, most of these broad theoretical concepts and constructs are of dynamic and exploratory nature and have been developed in a deductive process.
Popper (1976) argues that real progress in science and the discovery of new things are carried out through deduction. While induction can lead to a detailed and precise description and classification of the phenomena under study, in principle it cannot lead to any explanation. In turn, using deduction, one can propose mechanisms explaining not only how things look, but also why they look the way they do. In this approach, the researcher proposes a theoretical model, which may initially come across as surprising, especially if it connects things not ostensibly related or similar to each other. However, if the model is correctly developed and empirically testable, it can considerably enhance our understanding of human behavior, by going beyond the observable facts and offering new insights. Some of the dual constructs in Table 1 are examples of such deductive models explaining some phenomena. However, due to their disparate nature, not all of those constructs are readily amenable to testing, and some have yet to be operationalized. Moreover, their number is very large and they concern very specific domains of human psychology. Therefore, of great value is the opportunity offered by the TFM to integrate these dynamic and explanatory theoretical constructs with empirically established knowledge of the personality trait structure. Indeed, the two broad personality factors could play the role of a common-denominator or umbrella model, to use the term coined by Saucier et al. (2014), by linking dispositional personality traits with theoretical constructs and models, and the latter with each other (Digman 1997; Saucier et al. 2014; Strus and Cieciuch 2017). Some efforts have been already made to achieve such integration. Below, we present two current models striving to integrate knowledge about personality in which the TFM plays a crucial role.
The Big Two in the Hierarchical Structure of Personality: The Cybernetic Big Five Theory
Recently, DeYoung (2015) has proposed a comprehensive, integrative, and mechanistic explanatory theory of personality drawing on cybernetics as a study of adaptive, goal-directed, and self-regulating systems. Similarly to the FFT, his Cybernetic Big Five Theory (CB5T) assumes that all constructs describing individual psychological differences can be divided into personality traits and characteristic adaptations. However, in the CB5T personality is understood as an evolved cybernetic system, and so traits are taken to reflect variation in the parameters of evolved cybernetic mechanisms, while characteristic adaptations are thought to represent relatively stable goals (desired future states), interpretations (factual and evaluative representations of the current – past and present – state of the world and the self), and strategies (plans, actions, skills, and routines used in attempting to transform the current state into the desired future state) specified in relation to the individual’s particular life circumstances.
Due to the above considerations, the CB5T adopted a three-level hierarchical structure of traits. As the name of the theory indicates, this taxonomy is based on the Big Five factors treated as the basic cybernetic parameters and regarded as the main dimensions of personality trait covariation. The CB5T recognizes the cybernetic function of each of the Big Five factors and their 10 aspects, because CB5T additionally includes a set of 10 intermediate factors or traits placed in the personality structure between the basic domains (the Big Five) and their facets. Importantly, this 10-aspect substructure of the Big Five was established empirically (see DeYoung 2015), in contrast to the facets, which mostly were speculatively derived, and there is no widespread consensus as to their number or content. For this reason, the CB5T is based on the Big Five domains and their aspects, whereas the number and identity of the facets is deemed unknown.
Both the Big Five traits and their aspects fulfill specific cybernetic functions as parameters of the cybernetic personality system, although they are linked to different processes and stages of the cybernetic cycle. These functions may characterize people’s ongoing cybernetic adjustments to their environments in terms of selection among the existing goals and strategies. However, a cybernetic theory of human behavior needs to explain the process by which new goals and strategies are created, thus transforming the individual’s collection of characteristic adaptations. Thus, the CB5T accounts for that process via the two metatraits of stability and plasticity, which occupy the top level of personality trait structure. While cybernetic systems must maintain stable operation, they must also allow sufficient plasticity to adapt to unpredictable environments and change.
Stability reflects the capacity of the cybernetic system to resist disruption and replacement of its operative goal with immediate goals (e.g., expressing anger or pursuing a distraction) which interfere with longer-term goals when faced with an anomaly. Then, stability is related to the resistance of characteristic adaptations to interruption by emotions, impulses or doubts, and to descending into chaos under the influence of anomalies or threats. In turn, plasticity reflects the readiness of the cybernetic system to generate new characteristic adaptations voluntarily, in response to incentives (the reward value of the unknown), and not only as a result of stressors causing instability and disintegration. In the face of an anomaly, plasticity may provide a motivation to explore by putting the current operative plans on hold and formulating some new interpretation, strategy, or goal (DeYoung 2015).
In light of the above, the two metatraits of stability and plasticity represent the broadest psychological properties of the cybernetic system. The cybernetic function of stability (vs. instability) is the protection of goals, interpretations, and strategies from disruption by impulses, while that of plasticity (vs. rigidity) is exploration in terms of the creation of new goals, interpretations, and strategies. They both refer to whether an individual’s goals, interpretations, and strategies are stable or unstable, plastic or rigid in relation to entropy (chaos, the unknown, uncertainty, or unpredictability) as a fundamental problem for any cybernetic system.
Summing up, at the core of the CB5T theory is a hierarchical structure of personality traits built around the Big Five dimensions with a crucial role played by the metatraits derived from the TFM. The prominent function of the later can be traced to the nature of the theory itself, i.e., an interpretation of personality within the framework of cybernetics as a study of goal-directed, self-regulating systems.
The CB5T has implications for well-being and psychopathology linking them to function and dysfunction in traits and characteristic adaptations. Moreover, DeYoung (2015), comparing the advantages and limitations of the CB5T with other salient theories (e.g., social-cognitive theories), called the CB5T an integrative, or synthetic theory. However, it seems that the integrative potential of the CB5T does not lie in providing a platform for reconciling various models and constructs in their original form, but rather in reinterpreting them within the framework of an empirically established and theoretically interpreted (in mechanistic exploratory terms) model of personality traits and within a precisely conceptualized domain of characteristic adaptations (in contrast to the FFT), with a focus on the causal dynamics between traits and characteristic adaptations. Moreover, as noted by DeYoung (2015), the hierarchical model of personality traits is an oversimplification at levels below the Big Five, as the structure of personality is not simply organized. According to DeYoung (2015), the CB5T is compatible with the existence of these additional relationships, but it does not change the fact that some traits, aspects, and facets have cross-loadings, being composites of two or more higher-order traits.
The Big Two in a Circular Structure of Personality: The Circumplex of Personality Metatraits
Gamma/integration is a CPM reinterpretation of the GFP, which enables the inclusion of accurate findings and correct interpretations from GFP proponents (Musek 2007; Rushton and Irwing 2011) while avoiding the weaknesses and misinterpretations pointed out by its opponents (see Revelle and Wilt 2013). In the CPM model, the GFP (i.e., gamma/integration) retains its theoretical meaning, but is located at the same level of the circumplex structure as alpha/stability and beta/plasticity (cf. Revelle and Wilt 2013), rather than hierarchically above them. In other words, from the CPM perspective, most controversies linked to the GFP result from it being assigned the wrong status within the structure of personality. Gamma/integration is one of two possible combinations of alpha/stability and beta/plasticity (both high vs. both low), with the second combination being delta/self-restraint (high alpha/stability and low beta/plasticity vs. low alpha/stability and high beta/plasticity). The presence of delta/self-restraint results both from the logic of the circumplex structure already incorporating alpha, beta, and gamma (Revelle and Wilt 2013), as well as from empirical findings indicating different-sign correlations of alpha/stability and beta/plasticity with other variables (e.g., DeYoung et al. 2002; cf. Becker 1999).
Moreover, the CPM assumes that the opposite poles of each metatrait exhibit some psychological meaning beyond simple opposition, defining the positive and negative poles of alpha, beta, gamma, and delta separately (Strus et al. 2014), with the negative ones being labeled disinhibition, passiveness, disharmony, and sensation seeking, respectively. Therefore, the CPM proposes an octant structure of personality metatraits, including four bipolar metatraits (dimensions) or eight unipolar ones (octants). Each of the eight octants represents a certain configuration of the Big Five traits (see Fig. 3).
According to Strus et al. (2014), the addition of another two metatraits to alpha/stability and beta/plasticity in conjunction with the assumption of a circumplex structure of metatraits enhances the integrative potential of the TFM, providing foundations for a truly comprehensive, wide-ranging theoretical synthesis. The CPM model can be treated as a kind of matrix accommodating various constructs and models of personality, temperament, emotion, motivation, well-being, and psychopathology, of which many are in fact circular models (Strus and Cieciuch 2017). Moreover, the CPM makes it possible to reconcile the tradition of personality description in terms of traits with its description in terms of types (Strus et al. 2014), as the gamma and delta poles directly correspond to basic personality prototypes understood as the most common configurations of the Big Five personality profiles (Gramzow et al. 2004). At the least, the CPM model facilitates, more readily than TFM, consistent theoretical integration of the trait (dispositional) approach to personality with those personality theories that make use of dynamic and explanatory theoretical constructs. This is so because many of the constructs presented in Table 1 exhibit closer theoretical correspondence with gamma/integration and delta/self-restraint than with alpha/stability and beta/plasticity. Examples include Block’s constructs of ego-control and ego-resiliency, showing affinity to delta-plus and gamma-plus in terms of psychological meaning, respectively (cf. Gramzow et al. 2004). In turn, the activity of the neurobiological BIS and BAS systems (Gray 1991) as mechanisms explaining the basic dimensions of temperament (anxiety and impulsivity) appear to be related to gamma-minus and delta-minus, respectively (see Strus and Cieciuch 2017).
However, it is worth noting that to date the CPM has not offered (in contrast to the CB5T) any explanatory mechanisms underlying the metatraits. Instead, it has demonstrated its potential to provide a platform for a wide-ranging theoretical synthesis. The CPM can be deemed a bridge between the trait-dispositional and dynamic-exploratory components of personality (e.g., characteristic adaptations), as well as an “intermediator” or “broker” linking constructs from other models, showing connections, and enabling exploration of deeper interrelationships between the phenomena that they describe. The CPM may also provide inspiration for further development or refinement of some models (see Strus and Cieciuch 2017). Obviously, the CPM is quite similar to Becker’s model (1999) as both consist of eight unipolar octants whose psychological content is much alike. However, the meanings of the metatraits comprising the CPM and Becker’s model differ in some important aspects, due to the some essential differences between the two models particularly pronounced in terms of the theoretical origins, context, and implications.
A major problem in the psychology of personality and individual differences is the concurrence of numerous models of broader or narrower scope with unclear or unknown interrelationships. Indeed, this state of affairs brings to mind ancient Greece, where philosophers pondering the nature of the world sought arche, which was supposed to be the substance and origin of all existing things. Different schools of thought advanced different positions in this respect: arche was claimed to be water by Thales of Miletus, air by Anaximenes, fire by Heraclitus, and earth by Xenophanes. In turn, Empedocles believed that each of his predecessors was in part right and that the world consisted of four elements: water, air, earth, and fire. These elements could combine with each other creating an infinite number of arrangements as a result of the action of two forces: love and strife, also termed attraction and repulsion.
Actually, Empedocles’s reasoning contains several important characteristics, which have enabled the emergence of the TFM in the field of psychology: (1) differences between theoretical models should entail synthesis incorporating legitimate insights from various models, rather than the choice of one over others; (2) of particular value are simple solutions and models, especially those which provide a point of departure for more informative and sophisticated models; (3) a good model contains a description of both a static structure and dynamic forces which constitute an explanatory mechanism. The TFM, stemming from both psycholexical and questionnaire studies of five factors of personality, as well as from some other research paradigms, may integrate many theoretical insights and models (such as the dual constructs presented in Table 1) in a straightforward way, offering a good starting point for further, more sophisticated models describing both the static structure and dynamic forces of personality.
There is a considerable body of research indicating the existence of two broad and orthogonal factors located at the top of the personality structure (e.g., Chang et al. 2012; DeYoung 2006; DeYoung et al. 2002; Digman 1997). Moreover, many authors report evidence supporting their substantive nature (e.g., Chang et al. 2012; DeYoung 2006), i.e., that they are not (or not entirely) method factors or response biases, resulting from, e.g., egoistic and moralistic biases in self-perception (Paulhus and John 1998; see Table 1). Obviously, the TFM is not in itself sufficient for describing that structure and does not invalidate more informative models, such as the FFM or HEXACO (DeYoung 2015; Saucier 2008; Saucier et al. 2014). However, the broadest dimensions of personality appear to be truly cross-culturally universal, biologically based, and capable of wide-ranging theoretical integration, and as such form a parsimonious model with a unique theoretical potential (DeYoung 2015; Saucier et al. 2014; Strus et al. 2014; Strus and Cieciuch 2017).
Nevertheless, further research is needed to confirm the identity of the two broad factors established independently in questionnaire and psycholexical studies. They clearly correspond to each other and their identity is highly likely, but the results of existing investigations are not conclusive (De Raad et al. 2010; Saucier 2008; Saucier et al. 2014). For instance, the pattern of relationships of the Big Two factors with the Big Five ones differs to some extent (particularly in the case of neuroticism/emotional stability; cf. De Raad et al. 2010; Digman 1997; DeYoung 2006; Saucier et al. 2014), and both Big Two factors reveal some differences in theoretical meaning.
The relationship between the Big Two factors is likely to be clarified within the CPM model (Strus et al. 2014), which represents one of the two main avenues for further theoretical extension of the TFM. The CPM continues the line of research adopting a circumplex structure of personality traits (Becker 1999; Eysenck and Eysenck 1985; Wiggins and Trapnell 1996), and supplements the TFM with two additional metatraits. In this way, the CPM enhances the theoretical synthesis potential of the TFM, forming a matrix or platform for accommodating different constructs and models in their original form, and revealing their internal and external interrelationships (Strus and Cieciuch 2017). However, to date the CPM has not offered any explanatory mechanisms underlying the metatraits.
The second avenue for theoretical extension of the TFM, the CB5T, does not include the GFP or a fourth metatrait (delta). However, it does fully retain the hierarchical model of personality structure and accommodate the empirically derived Big Five aspects, which are absent in direct form from the CPM. Most importantly, the CB5T offers mechanistic explanations of the Big Two factors of personality, as well as descriptions of causal dynamics between traits and characteristic adaptations, although its integrative potential involves reinterpretation of other personality constructs and models within its conceptual frame, rather than enabling integration in their original form.
The work was prepared within Grants 2014/14/M/HS6/00919 from the National Science Centre, Poland.
- Bakan, D. (1966). The duality of human existence. An essay on psychology and religion. Chicago: Rand Mcnally.Google Scholar
- Block, J. H., & Block, J. (1980). The role of ego-control and ego-resiliency in the organization of behavior. In W. A. Collins (Ed.), Development of cognition, affect, and social relations (Vol. 13, pp. 39–101). Mahwah: Erlbaum.Google Scholar
- De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlacic, B., Di Blas, L., Hrebickova, M., Szirmak, Z., Szarota, P., Perugini, M., Church, A. T., & Katigbak, M. S. (2010). Only three factors of personality description are fully replicable across languages: A comparison of 14 trait taxonomies. Journal of Personality and Social Psychology, 98, 160–173. doi: 10.1037/a0017184.CrossRefPubMedGoogle Scholar
- Gray, J. A. (1991). The neuropsychology of temperament. In J. Strelau & A. Angleitner (Eds.), Exploration in temperament: International perspectives on theory and measurement (pp. 102–128). New York: Plenum Press.Google Scholar
- McAdams, D. P. (1988). Power, intimacy, and the life story: Personological inquiries into identity. New York: The Guilford Press.Google Scholar
- Popper, K. R. (1976). The myth of the framework. In E. Freeman (Ed.), The abdication of philosophy – Philosophy and the public good: Essays in honor of Paul Arthur Schilpp (pp. 23–48). LaSalle, IL: Open Court.Google Scholar
- Rushton, J. P., & Irwing, P. (2011). The general factor of personality: Normal and abnormal. In T. Chamorro-Premuzic, S. von Stumm, & A. Furnham (Eds.), The Wiley-Blackwell handbook of individual differences (pp. 134–163). London: Blackwell Publishing.Google Scholar
- Saucier, G. (2008). Measures of the personality factors found recurrently in human lexicons. In G. J. Boyle, G. Matthews, & D. H. Saklofske (Eds.), The sage handbook of personality theory and assessment (Vol. 2, pp. 29–54). Los Angeles: Sage.Google Scholar
- Saucier, G., Thalmayer, A. G., Payne, D. L., Carlson, R., Sanogo, L., Ole-Kotikash, L., Church, A. T., Katigbak, M. S., Somer, O., Szarota, P., Szirmak, Z., & Zhou, H. (2014). A basic bivariate structure of personality attributes evident across nine languages. Journal of Personality, 82(1), 1–14. doi: 10.1111/jopy.12028.CrossRefPubMedGoogle Scholar
- Schwartz, S. H., Cieciuch, J., Vecchione, M., Davidov, E., Fischer, R., Beierlein, C., Ramos, A., Verkasalo, M., Lönnqvist, J.-E., Demirutku, K., Dirilen-Gumus, O., & Konty, M. (2012). Refining the theory of basic individual values. Journal of Personality and Social Psychology, 103, 663–688. doi: 10.1037/a0029393.CrossRefPubMedGoogle Scholar
- Strus, W., & Cieciuch, J. (2017). Towards a synthesis of personality, temperament, motivation, emotion and mental health models within the Circumplex of Personality Metatraits. Journal of Research in Personality, 66, 70–95. doi: 10.1016/j.jrp.2016.12.002.
- Wiggins, J. S., & Trapnell, P. D. (1996). A dyadic – interactional perspective on the Five Factor Model. In J. S. Wiggins (Ed.), The Five-Factor Model of personality (pp. 88–162). New York: The Guilford Press.Google Scholar