|
nif:isString
|
-
Thirty-two right-handed undergraduate psychology students (gender: 21 female, 11 male; age: M = 23.3 years, SD = 4.7) took part in our study in exchange for course credit points. All participants gave written informed consent and could decide to discontinue participation at any time. The research design was approved by the Ethics Committee of the German Psychological Society and performed in agreement with the Declaration of Helsinki.
Each participant played the following card game with the computer. First, the participant sees a screen with ten cards face down and receives the information that each card has a number from 1 to 10 and each number exists only once. Then, the participant draws one card. While choosing a card, an arrow in the middle of the screen indicates whether the higher or the lower card wins. Once the participant has chosen a card, that card is immediately turned over. Finally, a computer-drawn card from the remaining nine follows 7 s later (Fig. 1). While winning has no consequences, losing leads to an electric shock with a 50% probability. This design is related to Preuschoff et al. [19]. We modified the general procedure with respect to two different aspects. Firstly, instead of betting on a higher or lower second card with both cards drawn by the computer, our participants choose the first card out of ten themselves under a given bet. This modification was made to keep motivation levels sufficiently high by giving participants more freedom and the option to develop hypotheses about any existent general rules (which did not exist). Secondly, instead of losing or winning money, participants received electric shocks if they lost a bet. The idea of giving a shock only in some of the lost cases is due to a general habituation effect. As we work on the fear of electric shock rather than on the shock itself, we decided that the shock would be given with a 50% probability. The type of the bet changes in each round.
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0120989.g001 Experimental design. This approach, which excludes any form of active decision-making processes in the period after the first card, enables us to analyze pure sensitivity towards the likelihood of unpleasant events which is not contaminated by active decision-making processes. Showing the first card implicitly reveals the probability of receiving an electric shock associated with this card. After revealing the first card, we measure SCRs until the second card is turned over. In order to ensure that participants pay attention, we subsequently ask them to indicate whether they won or lost the bet. Incorrect answers were also followed by an electric shock. The software package Psychtoolbox-3 (www.psychtoolbox.org) running on MATLAB 7.1 (MathWorks Inc., United States) was used for stimulus presentation and response acquisition. Participants were firstly instructed in written form and then verbally. In addition, each participant performed a practice run of 2 rounds without electric stimulation, in order to understand the task instructions. Afterwards, each participant played a total of 60 rounds with electric stimulation. In order to cover the full range of possible cases, the sequence of the first cards was arranged in a way that allowed each possible outcome (combination of the bet type and the first card) to occur exactly six times in one session. The order of this sequence and the value of the second card were random. The average duration of each session was 45 minutes (min).
Pain was delivered via electric stimulation with a Canicom 800 (Num’Axes, France) (Levels of stimulation: 15; frequency: 1.023 Hz; duration: 115 ms, power range: 0.5 to 206 mW). The stimulation was administered via two flat AG-AGCL electrodes of 10 mm diameter being placed at the medial phalanges of the digits II and III of the dominant hand. Individual levels of pain stimulation were calibrated using a standard procedure: Shocks were presented in an ascending series of intensity until the participant indicated that the shock was painful. Once a painful level was reached, the previous stimulation level was used during the experiment.
We used a 16-channel bioamplifier (Nexus-16; Mind Media B.V., the Netherlands) and the corresponding recording software Biotrace (Mind Media B.V.) to record electrodermal responses. Unfiltered SC data were acquired using a customer-specific SC sensor. The sensor maintained a voltage of less than 0.8V between the two flat AG-AGCL 10 mm diameter electrodes, which were placed at the medial phalanges of the digits II and III of the non-dominant hand. In accordance with common recomendations [20], the electrode sites were prepared with an isotonic paste (TD-246, Discount Disposables) and there was a 5 min pause between attaching the electrodes and starting recording. SC data were sampled at 32 Hz.
Skin conductance data was analyzed using Ledalab (www.ledalab.de) applying continuous decomposition analysis to disentangle phasic components from tonic activity [21]. The integrated skin conductance response (ISCR), which is defined as the time integral of the phasic driver for a relevant time interval, was used as a measure for the phasic electrodermal response to a given stimulus. In order to account for the typical skewed distribution of magnitude of electrodermal responses, individual ISCRs were standardized by the formula: ISCR = log(1 + ∣ISCR∣) × sign(ISCR) [22], [23]. ISCRs for the time interval of 1-3 s after flipping the second card were used to analyze the effect of pain stimulation. The delay in SCRs has been reported in the literature [24]. In order to measure sensitivity towards the likelihood of receiving an electric shock, ISCRs were computed for the 5 s window starting 2 s after turning the first card. We skip the first two seconds in order to discard the effect of motor activation due to card selection (compare Fig. 2).
Figure data removed from full text. Figure identifier and caption: 10.1371/journal.pone.0120989.g002 Phasic (driver) information after the first card for three cases (8 s).Shaded areas indicate the within-participant standard errors of the mean. For ANOVA analyses, degrees of freedom were corrected by means of the Greenhouse-Geisser method were necessary and Bonferroni post-hoc tests were used for pair-wise comparisons of means.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}