Humboldt-Universität zu Berlin - Faculty of Life Sciences - General Psychology


The acquisition of action-effect bindings and their functional role in action control

Since November 2001
Principal Investigators: Dr. Dieter Nattkemper, Prof. Dr. Peter A. Frensch
Supported by the Deutsche Forschungsgemeinschaft (DFG)

In the tradition of ideomotor approaches to the control of voluntary actions contemporary approaches suggest that the relations between actions and the sensory events they effectuate are acquired and represented via action-effect binding whenever movements contigently produce particular environmental effects. Numerous experimental observations from the last two decades have provided convincing evidence that these action-effect bindings actually play a functional role in action planning. This research project, firstly, aims at elucidating the nature of the mechanisms underlying action-effect binding. Some controversy has centered around this issue: Proponents of the associative view of action-effect binding hold that the acquisition of action-effect structures is the result of automatically operating associative mechanisms whereas others suggest that action-effect representations result from intentional mechanisms that bind actions to anticipations of the external events that they produce. The second aim of our research project is to elucidate the functional role that action-effect bindings play in action planning. In the tradition of the ideomotor hypothesis, it is widely accepted that effect codes are functional in response selection. According to this view, activating the representation of an action-effect automatically activates the associated motor pattern that has been learned to produce the given effect. However, this view is challenged by observations indicating that effect representations do not play a functional role in the initial phase of action preparation but, rather, come into play after response selection in the further course of action planning.   

Further Readings:
  • Nattkemper, D., & Frensch, P. A.  (2004). Action-effect binding in voluntary action control – Empirical facts and open questions. In A. Mecklinger, H. Zimmer & U. Lindenberger (Eds.), Bound in memory: Insights from behavioral and neuropsychological research. Aachen: Shaker Verlag.

Implementing Verbal Task Instructions within the Cognitive System

Since 2001
Principal Investigator: Dr. Dorit Wenke

The question underlying this research project is how exactly verbal task instructions are translated into, and are used to control behaviour. That is, how are task sets or cognitive control structures that are assumed to guide task performance derived from verbal instructions?
This question was motivated by the observation that humans– at least those with relatively mature and intact frontal lobes – can follow instructions quite effortlessly even in situations that are completely new and arbitrary. We do not need weeks and months of operant conditioning as animals would in order to perform comparable tasks in a way we expect them to.
However, although instruction following is such an outstanding human ability, we know surprisingly little about the processes and representational structures that underlie this ability. In this research project, we have been concerned with (a) how the contents (i.e., the wording of an instructions and the examples provided) are used to configure task sets, (b) how fast and by which processes task sets are implemented in the cognitive system, and (c) how initial task instructions and task practice affect learning.

Further Readings:
  • Wenke, D., & Frensch, P. A. (2005). The influence of task instructions on action coding: Constraint setting or direct coding?. Journal of Experimental Psychology: Human Perception and Performance, 31, 803-819.
  • Wenke, D., Gaschler, R., & Nattkemper, D. (2005). Instruction-induced feature-binding. Psychological Research, published online at: .

Mechanisms of information reduction during skill acquisition

Since 1.1.2006
Principal Investigator: Prof. Peter A. Frensch
Researcher: Géza Harsányi
Supported by the Deutsche Forschungsgemeinschaft (DFG)

Starting point of the research project are two empirical results concerning information reduction in cognitive skill acquisition: (1) Humans learn, with training, to limit their task processing to relevant task information only, and (2) The extent of this information reduction can be influenced via speed and accuracy instructions. Both findings cannot yet be explained convincingly; rather, two competing classes of theories both offer plausible explanations. One class of theory is based exclusively on the operation of data driven learning mechanisms, and another class of theory assumes an interaction between data driven and top-down learning mechanisms. The global goal of the research project is to answer two questions: (1) how do top-down control processes affect information reduction? (2) how do speed-accuracy instructions influence information reduction? The answers to these questions will reveal fundamental characteristics of human learning. In addition, findings on the interplay of data-driven learning mechanisms and top-down activated control processes will allow us to understand what bottom-up and top-down processes contribute to the control of human action.

Further Readings:
  • Haider, H., Frensch, P. A., & Joram, D.  (in press). Are strategy shifts caused by data-driven processes or by intentional processes? Consciousness & Cognition.

The functions of conflict for information reduction during skill acquisition

Since 1.7.2006
Principal investigator: P.A. Frensch
Researchers: N.N.
Supported by the Deutsche Forschungsgemeinschaft (DFG)

Empirical findings have shown that humans can learn, with practice, to differentiate between relevant and irrelevant task information and to limit their task processing to relevant information only. Our theoretical framework (Haider-Frensch Unexpected-Event Hypothesis) explains information reduction as the result of an interaction between obligatory data-driven processes, on the one hand, and top-down activated, strategic control processes, on the other hand. According to the model, data driven learning mechanisms can lead to behavior that conflicts with expected behavior. Conflicts between expected and actual behavior can trigger top-down influenced attributional processes (e.g., hypothesis testing) that search for causes of the experienced expectation violation and can lead to detection of irrelevant task information. Detection of irrelevant task information can then foster the generation of a more efficient task strategy. In the research project we investigate under which conditions conflicts between expected and experienced behavior are noticed and lead to information reduction, that is, to a change in the task strategy used.

Further Readings:
  • Frensch, P. A., Haider, H., Rünger, D., Neugebauer, U., Voigt, S., & Werg, J. (2003). Verbal report of incidentally experienced environmental regularity: The route from implicit learning to verbal expression of what has been learned. In L. Jiménez (Ed.), Attention and implicit learning. New York: John Benjamins Publishing Company. (p. 335-366).

Executive control in multi-tasking situations

Principal Investigator: Dr. Torsten Schubert

Main Goal
We investigate how putative processing limitations of the cognitive system influence the stream of processing operations when humans perform two or more actions simultaneously. Which processing operations are subjected to capacity limitations and, in addition, how humans cope with these capacity limitations during task performance? In particular, we are interested in uncovering the psychological mechanisms as well as the neural implementation of executive control in multi-tasking situations. For that purpose, we use research methods from experimental psychology, e.g. reaction times and errors, as well as methods from modern cognitive neuroscience, e.g. fMRI. In addition, we investigate patients with different neurological disorders. For a list of separate projects see below.
Dual-task situations and non-conscious information processing
Principal Investigator: Dr. Torsten Schubert
Researchers: Christine Stelzel, Marina Palazova, alumni Dr. Rico Fischer
Supported by a DFG grant to Torsten Schubert (2001-2005)

Previous studies usually investigated how human performance suffers from bottlenecks in dual-task situations. Differently to that our research interest is directed to the mechanisms allowing humans to achieve performance goals despite these bottlenecks.

In the present project we investigate whether non-consciously triggered information processing may circumvent a bottleneck in dual-task situations. A number of studies has suggested that non-consciously presented sensory information may activate directly associated motor, semantic, and emotional memory contents, thereby circumventig resource-demanding processing stages. It is an open question whether that kind of information may be processed despite the involvment of a capacity limitation in dual tasks. For our research, we developed a new paradigm that allows to assess the impact of non-conscious information processing on the performance of participants in the paradigm of the psychological refractory period. Our main questions are: Can non-consciously triggered information bypass a bottleneck between tasks and which factors influence the efficiency of this kind of processing in dual tasks? The investigations cover a wide range of studies where we investigate non-conscious information processing in the context of dual-task but also in the context of single-task situations.

Further Readings:
  • Schubert, T. (1999). Processing Differences Between Simple and Choice Reactions Affect Bottleneck Localization in Overlapping Tasks. Journal of Experimental Psychology: Human, Perception, & Performance, Vol. 25, 2, 1-18.
  • Fischer, R., Schubert, T., & Liepelt, R. (2006). Accessory stimuli modulate effects of non-conscious priming. Perception & Psychophysics, in press.

Learning in dual-task situations

Principal Investigator: Dr. Torsten Schubert
Researchers: Tilo Strobach, Dr. Peter Frensch, alumni Roman Liepelt
Supported by DFG grants to Torsten Schubert and Peter Frensch (2001-2007)

The observation of performance costs during the performance of two simultaneous tasks compared to single tasks, points to the existence of structural capacity limitations of the cognitive system. Recent studies, however, indicated that these performance costs may disappear after very extensive and prolonged dual-task practice. This finding has important implications for theories about the architecture of the cognitive system and, in addition, practical implications for the issue of skill acquisition in resource demanding situations. The main goal of the current project is to specify the nature of the learning processes leading to the disappearance of performance costs in dual-task situations. Two alternative hypotheses are investigated experimentally: According to the first hypothesis, participants acquire general executive knowledge on how to perfectly schedule (i.e., without any costs) the processing streams of two tasks. According to the second hypothesis, an improvement of the information transition between input and output information within the single-component tasks is responsible for the practice dependent improvement of the dual-task performance.

Localisation of executive functions in the brain with fMRI

Principal Investigator: Dr. Torsten Schubert
Researchers: Christine Stelzel, alumni Dr. Andre Szameitat, Dr. D.Y von Cramon
Supported by a DFG grant to Torsten Schubert, a grant of the Sommerfeld Foundation to Christine Stelzel, and by the Max-Planck Society.

Lesion studies with humans and monkeys as well as neuroimaging studies suggest that the lateral prefrontal cortex (lPFC) is involved in the control of goal-direct behaviour behaviour. The aim of the present project is to investigate the functional role of the lPFC for dual-task processing. In a series of own fMRI studies we showed that bilateral regions surrounding the inferior frontal sulcus (IFS), a subregion of the lPFC, are involved when participants perform dual tasks as compared to single tasks. In further studies we uncovered the specific functional role of that region. These studies showed: 1. that the IFS is involved indendently on the presented input and output information during dual-task processing, 2. the dual-task related IFS activation is associated with control processes regulating the temporal order of the processed tasks, 3. the localisation of the latter control function is distinct from regions reflecting increased demands on working memory load during dual-task compared to single-task processing. While these findings suggest a basic role of regions surrounding the IFS for dual-task processing, they leave open the question by which specific neural mechanisms the lPFC controls participants’ performance in dual tasks. A series of actual fMRI experiments is aimed to uncover that question.
Schubert, T. & Szameitat, A. (2003). Funtional neuroanatomy of interference in overlapping dual tasks: a study with fMRI. Cognitive Brain Research, 17, 733-746.

Further Readings:
  • Stelzel, C., Schumacher, E.H., Schubert, T., & D’Esposito, M. (2005). The neural effect of stimulus-response modality on dual-task performance – a study with fMRI. Psychological Research, Epub ahead.

Acquisition and forming of affective evaluations and their impact on judgements and behaviour

Supervisor: Prof. Dr. Peter A. Frensch
Doctoral Student: Dipl.-Psych. André Weinreich
LIFE - International Max Planck Research School

Emotions in terms of evaluative information processing are basic processes of the human mind. Affective Evaluations are formed the very moment a stimulus is perceived (Niedenthal, 1990; Zajonc, 1980) and this happens automatically (Bargh; 1996). They consitute the main component of attitudes and preferences and interact with psychological processes that act in parallel or subsequently (i.e. motivation, attention). We are mostly interested in the acquisition of affective evaluations and their influence on judgements, decisions and (consumer) behaviour. We aim at improving and expanding the existing data and models for the generation and acquisition of affective evaluations and their influence on and interaction with other psychological processes by means of reaction time (e.g. Fazio et al., 1986), ratings (e.g. Niedenthal, 1990) and psychophysiological methods of emotional priming. Beyond the perspective of general psychology we ask these questions in a developmental context. We hope to contribute to answering the following questions:
  • Under which circumstances does (un)conscious information processing of emotion influence  judgements and behaviour?
  • What are the basic mechanisms?
  • Assuming that people gain more expertise with age and at the same time undergo a decline in cognitive/deliberative processing resources, how does the impact of an automatic process like emotional information processing on judgements and behaviour change with age?
  • Is the influence of emotional information on judgements and behaviour stronger for older people than it is for younger? What could be a reason for that?
  • How are affective evaluations generated and object-specifically acquired?
  • Is there any reason why the age groups should behave differently?

Impairments of executive functions in patients with neurological disorders and in old aged persons

Principal investigator: Dr. Torsten Schubert
Researchers: Franziska Plessow (student), alumni Dr. Grit Hein, D.Y. von Cramon

Previous studies revealed severe dual-task difficulties in the above mentioned groups of persons compard to normal subjects. However, these studies used rather complex behavioural measures and paradigms and, therefore, left open the question which specific part of the dual-task processing stream is impaired in which group of subjects. In the current project, we aim to specify the dual-task difficulties of neurological patients and old-aged subjects. For that purpose, we use the well-defined paradigm of the Psychological Refractory Period in combination with the so-called locus-of-slack technique. The application of that research technique allows distinguishing between processing difficulties at the input, central, and/ or output stages of a dual-task situation. A first series of experiments revealed strong difficulties in dual-task input processing in patients with closed-head injury (CHI) and old-aged persons compared to age-matched and young controls. Further analyses indicated that a loss of inhibitory functioning leading to impaired task-order control seems to be responsible for the observed impairement of dual-task input processing in these subjects. A series of actually ongoing studies is aimed at testing specific dual-task difficulties in other groups of neurological patients, i.e., patients with Parkinson’s disease (see below).
Hein, G., Schubert, T., & von Cramon, D.Y. (2005). Closed head injury and input processing in dual-task situations. Experimental Brain Research,160, 223-34.

Further Readings:
  • Hein, G. & Schubert, T. (2004). Aging and Input Processing in Dual-Task Situations. Psychology and Aging, 19, 416 – 432.

Deep brain stimulation of the basal ganglia in patients with Parkinson's disease

Principal Investigators: Dr. Torsten Schubert and Dr. Jens Volkmann (Christian-Albrechts-University Kiel)
Researchers:Franziska Plessow (student)

In the present project, we use Deep Brain Stimulation (DBS) of the basal ganglia to investigate the dependence of various motor and cognitive processes on the functional state of the basal ganglia. Deep Brain Stimulation (DBS) is a recently introduced method for the treatment of the basal ganglia dysfunction – Parkinson’s disease (PD). In DBS an electrode is implanted in the GPi (or other structures of the basal ganglia). A permanent high frequency stimulation of the GPi through this electrode results in an improvement of information flow through the impaired basal ganglia-thalamo-cortical circuits in PD. By manipulating DBS some of the Basal ganglia nuclei (i.e., Gpi or STN) can be switched on and off. One can infer the causal role of the GPi in some cognitive or motor function by comparing the influence of the DBS on the performance of PD patients in some certain experimental task.

Further Readings:
  • Schubert, T., Volkmann, J., Müller, U., Sturm, V., Voges, J., Freund, H.-J. & von Cramon, D.Y. (2002). Effects of pallidal deep brain stimulation and levodopa treatment on reaction-time performance in Parkinson’s disease. Experimental Brain Research, 144, 8-16.