Major Research Interests / Research Projects
Research in our department investigates processes of cognition, affect/emotion, learning, and motivation from an action perspective on human functioning. According to this view, micro-processes of psychological functioning (e.g., perception; attention; reasoning; conditioning; memory) are tuned to the demands of "superordinate", personal levels of functioning (goal pursuit and goal adjustment; coping with threats and challenges; stabilization of self-concept and personal identity). In turn, "subordinate" levels of functioning are important for the regulation of behavior and can also have a backward effect on personal orientations and goal pursuit.
To investigate these questions, we make use of paradigms that were developed in Cognitive Psychology and Social Cognition research (e.g., semantic, affective, negative priming; task switching; evaluative conditioning; Stroop and Simon paradigms, IAT). Current research focuses on the following topics: (a) Action-regulation in affective processing (effects of opportunites and threats on the processing of positive and negative stimuli), (b) emotional action regulation (effects of valent stimuli on approach/avoidance behavior; moderating conditions and mediating mechanisms), (c) the self in motivation and cognition (the self as a source of personal motivation; self-serving biases in cognition, evaluation, and perception), (d) basic mechanisms of cognitive functioning (spreading activation, selective attention, inhibition, binding of features and responses, episodic memory retrieval), (e) theories and measures of implicit cognition (automatic activation of evaluations and associations), (f) cognitive representations of mental categories (automatic stereotype activation; moderating mechanisms; changing stereotypes).
Action and Cognition
Goal orientations and processes of action regulation influence – and are in turn influenced by – micromechanisms of cognition and information processing (Rothermund, in press-a; in press-b; 1998; see also Brandtstädter & Rothermund, 2002a, 2002b). The core hypothesis of our research is that information processing is tuned to what is relevant for action regulation and goal pursuit. As long as action resources are available for active goal pursuit (i.e., in an assimilative mode of action regulation), information processing is focused on the current goals and the means towards achieving the goals. When action resources become scarce and attempts to reach a goal are chronically frustrated (i.e., in an accommodative mode of action regulation), cognitive resources are withdrawn from blocked goals and redirected towards other, more promising goals and projects.
Previous experiments examined the relation between assimilative and accommodative modes of coping and cognitive processes relating to selective attention (Rothermund, 1998, 2003), perception (Rothermund & Brandtstädter, 1998b; Rothermund, Brandtstädter, Meiniger, & Anton, 2002), and motivated reasoning (Rothermund, Bak & Brandtstädter, 2005).
Life-span Developmental Psychology
(Klaus Rothermund, Anna Kornadt)
Research on life-span development in our department investigates factors that have an influence on adaptation to the challenges of the aging process, mechanisms of developmental regulation and successful aging. This includes experimental as well as questionnaire-based studies with focus on different topics , such as:
- (Context specific) age stereotypes and subjective well-being (e.g.,Casper, Rothermund, & Wentura, in press; Kornadt & Rothermund, 2011, 2012; Rothermund, 2005; Rothermund & Brandtstädter, 2003)
- Age discrimination (Rothermund & Mayer, 2009; Rothermund & Temming, 2010)
- Depression in later life (Rothermund & Brandtstädter, 2003)
- Adaptive changes in goals and values (Brandtstädter, Rothermund, Kranz, & Kühn, 2010 )
- Control beliefs in later life (Brandtstädter & Rothermund, 1994)
- Coping with age-related deficits and losses (Rothermund & Brandtstädter, 2003)
Most recently, we are involved in two successive interdisciplinary research projects on transitions in adulthood and perceptions of aging and older persons (2008 - 2012: "Zonen des Übergangs. Dimensionen und Deutungsmuster des Alterns bei jüngeren, älteren und alten Menschen"; 2012 - 2015: "Alter(n) als Zukunft. Zukunftsbezogenes Altershandeln in kulturvergleichender Perspektive") which are funded by the Volkswagen Foundation.
DFG-Research project on "Distractor-response bindings" (Distraktor-Reaktions-Bindungen)
(Klaus Rothermund, Carina Giesen)
Recent experimental evidence (e.g., Hommel, 1998; Logan, 1988) could show that relevant aspects of a stimulus become associated with response information and are encoded together in our memory as a unit (or "event file"). Importantly, whenever an element of such a stimulus-response (S-R) episode is repeated, the entire episode is retrieved, including the associated response, which can either facilitate or delay performance, depending on whether the retrieved response is appropriate or not.
In 2005, Klaus Rothermund, Dirk Wentura and Jan DeHouwer introduced their "stimulus-response retrieval" (SRR) theory and argued that such binding of stimulus and response information is not restricted to relevant stimuli, but extends to irrelevant stimuli (i.e., distractors) as well: Apparently, distractors are also integrated into such S-R episodes and may function as retrieval cues for the S-R episode (see also Frings, Rothermund, & Wentura, 2007).
Our research project aims to investigate the particular circumstances that foster or restrict the creation of such distractor-response bindings and influence the likelihood that a distractor is integrated into an S-R episode. In particular, we are interested in the interplay between processes of selective attention, feature binding and S-R retrieval in situations in which a relevant stimulus is accompanied by distractors.
The project is a co-operation together with Prof. Dr. Christian Frings and Birte Möller (University of Trier, Germany).
DFG-Research project on "Goal context and attention-control" (Zielkontext und Aufmerksamkeitssteuerung)
(Klaus Rothermund, Susanne Schwager)
Human behavior is characterized by an enormous variability and flexibility which is a necessary requirement for psychological functioning and adaptation. We believe that flexible self-regulation can not only be the result of deliberate choice but also of underlying automatic mechanisms (Rothermund, in press). Recent research sometimes provided evidence for a negativity bias in attention allocation (Pratto & John, 1991) and sometimes for a positivity bias (Kunda, 1990). Besides the heterogeneity of the results it does also not seem likely from a theoretical perspective that human information processing should be governed by a stable attention allocation which would lead to becoming locked up in motivational-emotional states. The counter-regulation theory recently suggested by Rothermund, Voss and Wentura (2008) provides an explanation for the variability of human behavior. In detail, we hypothesized that attentional biases are moderated by motivational phases during goal pursuit so that escalation or perseveration of affective-motivational states is prevented.
Support for this counter-regulation theory has been found in studies concerning the processing of valent information after positive or negative feedback (Derryberry, 1988, 1993; Rothermund, 2003; Rothermund, Gast, & Wentura, in press) and while anticipating future outcomes (Rothermund et al., 2008; Wentura, Voss, & Rothermund, 2009). Other aspects of human self regulation as for instance decision processes might be influenced by this mechanism as well.
DFG-Research project on "Context-dependent activation of stereotypes" (Kontextspezifische Stereotypaktivierung)
(Klaus Rothermund, Florian Müller)
According to textbook knowledge in Social Cognition, stereotypic attributes are automatically activated whenever social category information becomes salient. However, given that many categories are associated to a large number of different stereotypic attributes (e.g. old, wise, generous, fragile, lonely, ill, helpless, ...), a simultaneous activation of all of these attributes is quite unlikely. The crucial question thus is, which of these attributes become activated in a specific situation? In several priming studies, we tested a model of context-specific stereotype activation. The findings highlight that only those elements of a stereotype become activated that are relevant for the current context, or that fit with individuating information of the person with whom we interact (Casper, Rothermund, & Wentura, 2010, in press> ).
Measurement of Implicit Attitudes with the IAT
(Klaus Rothermund, Franziska Meißner)
The Implicit Association Test (IAT, Greenwald, McGhee & Schwartz, 1998) is one of the most widely used measures for implicit attitudes. Only a few studies, however, investigated the processes underlying the IAT effects. We try to approach this issue in three research lines:
- According to the association model, IAT effects represent automatic evaluations of categories (Greenwald et al., 1998). Our figure/ground model, however, assumes that IAT effects reflect salience asymmetries between categories. In a large number of experiments we found the IAT to depend on salience asymmetries, whereas evaluative associations did not have an influence independently of category salience (Rothermund & Wentura, 2001, 2004).
- It's likely that the to-be-measured associations are covered by recoding in most IATs. This kind of process simplifies the IAT task in the compatible block and thus influences the interpretation of compatibility effects. We therefore modified the IAT procedure so that compatible and incompatible response assignments switch randomly between trials within a single block. Recent experiments revealed that this IAT-RF ("IAT-recoding free") effectively prevents recoding and becomes immune against biased selections of stimuli (Rothermund, Teige-Mocigemba, Gast, & Wentura, 2009).
- Potentially, recoding processes could also contribute to the predictive validity of the IAT. With standard methods of analysis, this could hardly be examined because associations and recoding processes cannot be separated within one IAT. Therefore, we developed a multinomial processing tree model for the IAT in order to estimate different parameters for these cognitive processes. Preliminary findings confirm the validity of the model parameters. This sophisticated method could clarify some open questions regarding the underlying processes of IAT effects.
The priming methodology is one of the core research instruments for the investigation of cognitive processes and structures. Besides its original use to examine the mental representations of concepts and their mutual associations (e.g., Neely, 1991), the priming paradigm has also been used to investigate processes of selective attention and automatic memory retrieval ("Negative Priming"; e.g., Tipper, 2001). Recently, priming procedures have also become a fundamental tool in Social Cognition research to investigate the automatic activation of valence, attitudes, stereotypes and even motivational and behavioral tendencies (e.g., Bargh & Chartrand, 1999; Blair & Banaji, 1996; Dovidio, Evans, & Tyler, 1986; Fazio et al., 1986; Perdue & Gurtman, 1990; Rothermund, Wentura, & Brandtstädter, 1995).
Priming effects have typically been explained by a spreading of activation in a semantic network (e.g., Collins & Loftus, 1975). A closer examination of the mediating mechanisms has revealed, however, that priming effects are also strongly influenced by processes that are located at a peripheral stage (e.g., response facilitation and interference, judgmental tendencies). Dirk Wentura, Jan De Houwer and I have conducted some experiments that were designed to disentangle the processes underlying Affective Priming effects (De Houwer, Hermans, Rothermund, & Wentura, 2002; Eder, Leuthold, Rothermund, & Schweinberger, 2010; Rothermund & Wentura, 1998; Voss, Rothermund, Gast, & Wentura, 2010; Wentura & Rothermund, 2003). The upshot of these studies is that affective congruency effects are strongly influenced by interference between responses that are automatically triggered by the prime and the target. Affective priming effects thus should not be taken as direct evidence for a spreading of activation between affectively congruent nodes in an affective-semantic network (Bower, 1981).