Modulation of spatial and stimulus–response learning strategies by exogenous cortisol in healthy young women

https://doi.org/10.1016/j.psyneuen.2008.09.018Get rights and content

Summary

Glucocorticoids (GCs) are known to influence learning and memory processes. While most studies focus on the effects of GCs on the performance within a single memory system, we asked whether GCs modulate also the transition between hippocampus-dependent spatial and caudate nucleus-dependent stimulus–response memory systems. Eighty-four young healthy women received a placebo, 5 or 30 mg hydrocortisone orally. One hour later, participants were asked to locate a win-card in a 3D model of a room. The card could be located via two strategies: spatial (multiple distal cues) and stimulus–response (a single proximal cue). Relocation of the proximal cue after 12 trials revealed the strategy, number of trials to learning criterion the performance. As expected, more trials were needed to acquire the task with hydrocortisone. Remarkably, hydrocortisone switched the use of learning strategies towards more spatial learning (dose-dependently: placebo 4% < 5 mg 21% < 30 mg 32%), independent of autonomic and subjective arousal. The learning curves of spatial and stimulus–response learners were comparable. Our results demonstrate that exogenous GCs prior to learning affect the performance within a memory system and also coordinate the use of multiple memory systems. Taking into account this dual action of GCs will contribute to a better understanding of stress (hormone) effects on learning and memory.

Introduction

Glucocorticoids (GCs; cortisol in humans) secreted by the adrenal cortex regulate metabolic, immunological and cardiovascular homeostasis as well as cognitive functions, such as memory (Lupien and McEwen, 1997, Sapolsky et al., 2000, de Kloet et al., 2005). Effects of stress- or pharmacologically induced GC elevations on memory depend critically on the timing of GC activity. While GCs released around the time of learning enhance memory, memory performance is impaired when GCs are experienced out of the learning context (for a review: Joels et al., 2006).

Most studies that examined stress or GC effects on memory focused on changes in performance within a single memory system, mainly the hippocampus (Newcomer et al., 1994, Buchanan and Lovallo, 2001, Lupien et al., 2002, Abercrombie et al., 2003, Kuhlmann et al., 2005a, Roozendaal et al., 2006). However, it is important to note that memory is no unitary entity but consists of multiple anatomically and functionally distinct systems (White and McDonald, 2002, Squire, 2004). Two of these systems have been in the spotlight of the multiple memory systems literature: a hippocampus-dependent “cognitive” memory which has been associated with spatial learning and memory and a caudate nucleus-dependent “habit” memory that was related to stimulus–response (S–R) learning and memory (Packard and McGaugh, 1992, Kim et al., 2001, White and McDonald, 2002, Iaria et al., 2003, Bohbot et al., 2004). Though, both systems make distinct contributions to the optimization of behavior, they can interact both in a cooperative or competitive fashion (Poldrack and Packard, 2003, Voermans et al., 2004). This raises the question which factors determine in case of competition between memory systems the nature of interactions and the dominance of either system. Kim et al. (2001) suggested that stress plays a critical role in the modulation of multiple memory systems. They showed that stress prior to training in a water maze task led to a shift from predominant spatial to more S–R learning in rats. Similarly, Packard and Wingard (2004) reported that rats that were injected anxiogenic drugs predominantly displayed caudate nucleus-based S–R learning in a plus maze task, whereas vehicle-treated rats predominantly displayed hippocampus-based spatial learning. We translated these findings recently to humans and found that psychosocial stress modulated multiple memory systems in favor of caudate nucleus-dependent S–R learning and at the expense of hippocampus-dependent spatial learning in healthy men and women (Schwabe et al., 2007a). Moreover, we showed that S–R learning was most likely in the face of large cortisol increases. However, these increases in cortisol were confounded with other stress effects, such as autonomic and subjective arousal. Thus, this study allowed – same as the rodent studies cited above – no clear conclusion about the involvement of GCs in the modulation of spatial and S–R learning.

In the present study we examined whether the increase in cortisol is the mechanism underlying the stress-induced modulation of multiple memory systems. Therefore, 84 healthy young women were administered either a placebo or a low or high dose of hydrocortisone. Different doses of hydrocortisone were given because previous studies suggested that GC effects on memory are dose-dependent (Lupien and McEwen, 1997, Abercrombie et al., 2003). We hypothesized that hydrocortisone would shift learning strategies towards more S–R learning and that this effect would be most pronounced in the high hydrocortisone group. One hour after drug intake, participants completed a non-arousing learning task that was designed to differentiate spatial from S–R learning strategies in humans (Schwabe et al., 2007a). Subjects were presented a 3D model of a room and had to identify a “win-card” out of four that could be located with the help of a single proximal cue (S–R strategy) or the relation between multiple distal cues (spatial strategy). The applied strategy was inferred from the participants’ performance in a test trial in which the proximal cue was relocated as well as from their verbal report. To control for effects of autonomic and psychological arousal, heart rate and subjective feeling were measured at several time points across the experiment.

Section snippets

Participants

Eighty-four healthy women (University of Trier, Germany) participated in this study (mean age: 22.8 years, SD = 2.7 years; placebo group: 22.8 years, SD = 2.2 years; 5 mg hydrocortisone group: 22.5 years, SD = 3.2 years; 30 mg hydrocortisone group: 23.2 years, SD = 2.8 years; criteria: non-smoking, use of oral contraceptives (except use of Yasmin® and PettiBelle® which contain a moderate mineralocorticoid receptor antagonist), no reported history of psychiatric disorders or drug abuse). Participation was

Manipulation check

As expected, saliva cortisol concentrations differed significantly just prior to behavioral testing (all P < 0.001), while there were no significant baseline differences. Significant elevations in saliva cortisol concentrations were found in the 30 and 5 mg hydrocortisone groups (both P’s < 0.001) but not in the placebo group (P = 0.39; Table 1).

Strategy use

Cortisol treatment exerted a significant effect on the strategy used in the test trial (χ2(2) = 7.02, P < 0.03). While the spatial strategy was used by 32% of the

Discussion

The present study asked whether exogenous glucocorticoids in the absence of subjective and autonomic arousal are sufficient to modulate the use of spatial and S–R learning strategies. Indeed, participants that were administered hydrocortisone prior to training in a 3D spatial task changed their learning strategy and used – contrary to our expectation – significantly more often a spatial strategy than placebo-treated controls. This effect was dose-dependent. It is important, that the change of

Role of funding sources

Funding for this study was provided by the German Research Foundation (DFG, grant GRK 1389/1) the Netherlands Organisation for Scientific Research (NOW, grant NWO DN 95-420); the DFG and the NWO had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

Conflict of interest

All authors report no conflict of interest.

Acknowledgements

This work was part of the Trier-Leiden IRTG program, supported by grants DFG GRK 1389/1, NWO DN 95-420. We thank Nina Mariel Knapp and Linn Kühl for their assistance during data collection.

References (41)

Cited by (55)

  • Stress and prospective memory: What is the role of cortisol?

    2019, Neurobiology of Learning and Memory
    Citation Excerpt :

    In sum, the first goal of the present study was to investigate how cortisol affects PM performance by administration of 10 mg cortisol vs. placebo. This dose of cortisol mimics the physiological responses to stress, but crucially does not affect SNS activity (Schwabe, Oitzl, Richter, & Schächinger, 2009) and internal states that make participants aware of treatment (van Ast, Cornelisse, Meeter, Joels, & Kindt, 2013). Our second aim was to examine whether potential cortisol effects vary with regard to how strongly the PM task relies on prefrontal areas.

  • Stress and the trade-off between hippocampal and striatal memory

    2017, Current Opinion in Behavioral Sciences
View all citing articles on Scopus
View full text