In this experimental study we investigated the effect of state anxiety (induced by 7.5% carbon dioxide inhalation) on simple information processing. 

Participants (n=36) were a convenience sample of healthy volunteers. They attended one session and completed two 20-minute inhalations: one of 7.5% carbon dioxide enriched air, and one of medical air (i.e., placebo). Two tasks (audio-visual matching task and visual binary categorisation task) were completed during each inhalation.  Half of the stimuli in each task were degraded in order to examine whether the effects of anxiety interacted with stimulus clarity. To validate the anxiety manipulation, cardiovascular (blood pressure, heart rate) and subjective measures (anxiety, mood) were taken at baseline, and after the inhalations.

Data Extraction/Analysis: An independent researcher validated 20% of manual data entry (psychological and physiological readings at baseline, and after each inhalation). Input was found to be reliable (error rate 0%).

Audio-visual task: Task data were extracted from E-Prime 2.0 software (Psychology Software Tools, 2012) to SPSS Version 21.0 (IBM Corp., 2012) for analysis. Trials in which participants failed to respond within the 4000ms limit, were not analysed. These amounted to 0.2% of trials (19 out of 7200) in the air condition and 0.7% (53/7200) in the CO2 condition. Additionally, in terms of reaction time data, only correct responses were analysed, averaging across matching and unmatching conditions. 

Visual-binary task: Task data were extracted from Mousetracker using the built in Analyzer program (Freeman & Ambady, 2010). Trials with incorrect responses (4%), or where the initiation time was longer than 1,000 ms (1%) were not analysed. Responses were collapsed across category (Biological/Non-Biological) and both response sides were collapsed together by remapping all trajectories to the left onto the right side. The Analyser program outputs a .csv file with RT and AUC calculated. This .csv file was used to produce mean values for each dependent variable for each participant in each condition. These values were then input into SPSS 23 for inferential analysis.

Primary analyses were repeated measures ANOVA with gas (7.5% CO2, air) and stimulus clarity (clear, degraded) included as within-subjects factors.

The study protocol was published online prior to starting testing on the Open Science Framework (https://osf.io/gqs93/).

The data dictionary defines the contents of the data sheet comprehensively. This includes a description of the variables in data sheet, including units of measurement and coding where possible. 

Note on variable names: data relating to non-degraded stimuli presentation (both visual and auditory) are referred to as "clear" in the manuscript, but these variables are labelled as "fine" in the data sheet.

The deposit includes the following files:

2018/01/12 CO2 Information Processing Consent Form
2018/01/12 CO2 Information Processing Information Sheet
2018/01/12 CO2 Information Processing Data Dictionary
2018/01/12 CO2 Information Processing Data Sheet

Data files and data dictionary can be opened in Excel or SPSS software.