Elsevier

NeuroImage

Volume 20, Issue 2, October 2003, Pages 1026-1031
NeuroImage

Regular article
When the same response has different meanings:: recoding the response meaning in the lateral prefrontal cortex

https://doi.org/10.1016/S1053-8119(03)00357-4Get rights and content

Abstract

The ability to adapt our behavioral repertoire to different situations and tasks is crucial for our behavioral control. Since the same motor behavior can have different meanings in different task situations, we often have to change the meaning of our responses when we get into a different task context. In a functional MRI experiment we manipulated this response recoding process. Subjects were required to execute two simple spatial tasks in a task switching paradigm. In one condition both tasks required the same set of responses, hence each response had two different meanings depending on the relevant task (bivalent condition). In the other condition subjects used a separate set of responses for each task (univalent condition). While subjects were required to recode the meaning when switching from one task to the next in the bivalent condition, response recoding was not required in the univalent condition. We demonstrate that the lateral prefrontal cortex is involved in recoding of response meaning. These results extend previous assumptions on the role of the prefrontal cortex in behavioral control.

Introduction

One of the most fascinating properties of intelligent behavior is the human ability to apply a restricted behavioral repertoire to an infinite number of different task situations. Pressing a light switch, for example, can have different outcomes (switch the light on or off) depending on the context. To be able to use the same physical response (pressing the light switch) to achieve different goals (switching the light on or off) one has to recode the response meaning when switching from one task situation to the next Meiran, 2000, Schuch and Koch, 2003. We will refer to this process as “response recoding.” Recently it was found that neurons in the prefrontal cortex of monkeys represent the response meaning depending on a given task context (Asaad et al., 2000). In this study prefrontal cortex neurons were found to be activated in preparation of a specific response but only when the response occurred in a given task context. These results suggest that the prefrontal cortex might be involved in processing task-context specific response information (the response meaning).

In cognitive psychology the influence of the task context on task performance is usually investigated in so-called task-switching paradigms. A number of neuroimaging studies have used this paradigm to investigate task-related control processes Brass and von Cramon, 2002, Dove, 2000, Dove et al., 2000, Kimberg et al., 2000, Omori, 1999, Pollmann et al., 2000, Sohn et al., 2000. However, to our knowledge recoding the response meaning was never directly investigated with functional MRI. Some previous studies have manipulated the stimulus-response mapping (S-R mapping) between two tasks, which probably involves response recoding Dove et al., 2000, Nagahama, 2001, Pollmann et al., 2000. Reversing the S-R mapping yielded activation in the prefrontal cortex. However, because these studies did not manipulate response recoding independently from task switching, it is not clear whether the prefrontal activation was due to general task-set reconfiguration or to response recoding.

The aim of the present study was to directly investigate the cortical basis of our ability to use the same physical response to achieve different goals. To do so, we compared a situation in which subjects were required to recode the meaning of their responses while switching from one task to the next with a situation in which they were not required to recode the response meaning.

Section snippets

Experimental design

In the study we used a paradigm in which subjects had to switch between two simple spatial tasks (Meiran, 1996). In one task (task A) they were required to decide whether a square, which was presented in a two-by-two grid, was in the upper or lower half of the grid, and in the other task (task B) subjects had to decide whether the square was on the left or on the right side of the grid (Fig 1). The task executed in a given trial was indicated by two arrows presented either above or below the

Behavioral results

The behavioral data replicate previous studies; they show that subjects were significantly slower in switch trials (transition from task A to B or task B to A) than in repetition trials (repetition of task A or task B), F(1,17) = 82.3, P < 0.001. This switch effect (635 vs. 594 ms) reflects the costs of switching from one task to the next. This effect was larger in the bivalent (51 ms) than in the univalent condition (31 ms), F(1,17) = 12.2, P < 0.01, indicating, that the switch costs were

Discussion

By manipulating the overlap of response sets in the univalent and bivalent condition, we demonstrate that the LPFC mediates the change of response meaning when the person switched from one task to the next. An alternative explanation of this interpretation might be that participants, instead of recoding the response in switch trials, represent the response in a multidimensional response space. However, in incongruent trials this is not a reasonable strategy, since the same stimulus requires two

Acknowledgements

This work was supported by the German-Israeli Foundation for Scientific Research and Development (Grant Number: G-635-88.4/1999).

References (32)

  • M. Brass et al.

    The role of the frontal cortex in task preparation

    Cereb. Cortex

    (2002)
  • M.A. Burock et al.

    Randomized event-related experimental designs allow for extremely rapid presentation rates using functional MRI

    Neuroreport

    (1998)
  • Dove, A., 2000 Lokalisierung an internen Kontrollprozessen beteiligter Hirngebiete mit Hilfe des...
  • J. Duncan

    An adaptive coding model of neural function in prefrontal cortex

    Nat. Rev. Neurosci.

    (2001)
  • P.C. Fletcher et al.

    Frontal lobes and human memoryinsights from functional neuroimaging

    Brain

    (2002)
  • K.J. Friston

    Statistical parametric maps in functional imaginga general linear approach

    Hum. Brain Mapp.

    (1995)
  • Cited by (90)

    • What are we measuring when we measure task switch costs?

      2024, Current Opinion in Behavioral Sciences
    • Temporally and functionally distinct large-scale brain network dynamics supporting task switching

      2022, NeuroImage
      Citation Excerpt :

      Further studies using electrical stimulation mapping may be warranted to determine whether neural activation in the right precuneus region is essential for task switching. The visual-dorsolateral prefrontal network dynamics may be involved in executing appropriate stimulus-response translation in switch trials without interference from old rules (Brass et al., 2003; Koechlin et al., 2003; Sohn et al., 2007; Hyafil et al., 2009; Nee and D'Esposito, 2017). To the point, many fMRI studies of healthy adults reported that more intense hemodynamic activations were induced in the dorsolateral prefrontal regions when a task switch was required (Dove et al., 2000; Sohn et al., 2000; Braver et al., 2003; Badre and Wagner, 2006; Yeung et al., 2006; Karayanidis et al., 2010; Kim et al., 2012; Jamadar et al., 2015; Vallesi et al., 2015; Cole et al., 2016).

    View all citing articles on Scopus
    View full text