Data and documentation for the paper: Gaze in visual search is guided more efficiently by positive cues than by negative cues

2016-01-13T09:50:44Z (GMT) by Günter Kugler Bernard Marius 't Hart
<p>This is the data repository for the Article in PLoS ONE:</p><p><b>Gaze in Visual Search Is Guided More Efficiently by Positive Cues than by Negative Cues</b></p><p>Günter Kugler*, Bernard Marius ‘t Hart*, Stefan Kohlbecher, Wolfgang Einhäuser, Erich Schneider</p><p>*contributed equally</p><p></p><p>DOI: 10.1371/journal.pone.0145910</p><p>- - -</p><p>For clarification beyond what is covered in this text, please contact the corresponding author of the published manuscript.</p><p>All files are CSV files with numerical content in the form of a table with columns and rows. The first line contains text labels identifying the variable in each column. Within the subset of files for experiment 1 and experiment 2 (separable by the prefixes 'exp1_' and 'exp2_') the numbers identifying individual participants are consistent across files. Although the numbers 1-14 are used in all files, a different group of participants took part in each of the two experiments.</p><p><br></p><p>Short description of the experiments</p><p>------------------------------------</p><p><br></p><p>We investigate how an instruction (cue) that describes a feature that is not present in the seach target could affect search strategy.</p><p><br></p><p>For a more complete explanation, please refer to the published manuscript.</p><p><br></p><p>Experiment 1: The main experiment for this project, where participants were cued with a color that identified half the items in the search array. All items contained an object which could ba a foil or a target. The features identifying the target were chosen to promote serial fixation of items. These items could contain the target, so that in the most efficient search items with that color would have to be sought out (positive cues) or these items could not contain the target, so that in the most efficient search items with that color would have to be avoided (negative cues). We used arrays of 5 different item set sizes, and presented all 8 colors as both negative and positive cues in blocks of 10 searches, with targets present as well as absent, for a total of 160 searches per participant. The cues were always correct. Participants had to report if the target was absent or present. By recording eye-movements along with behavioral measures (correct replies and response time), we could investigate the search strategies that were induced by the two cue types.</p><p><br></p><p>Experiment 2: In experiment 1 there were always 3 non-cued colors. In this second experiment we control for the effect of the number of cued and non-cued colors, and also introduce a neutral cue. In most other respects the experiment is identical to experiment 1. Some smaller differences are described in the manuscript.</p><p><br></p><p>Files for experiment 1:</p><p>-----------------------</p><p><br></p><p>* exp1_correctAsnwers.csv</p><p>(Figure 3, correctness of key-press response)</p><p><br></p><p>- TargetAbsent: 1=correct reply in target absent version, 0=incorrect reply in target absent version</p><p>- TargetPresent: 1=correct reply in target present version, 0=incorrect reply in target present version</p><p>- Condition: 1=positive cue, 2=negative cue</p><p>- ArraySize: number of items in the search display [6,12,18,24 or 30]</p><p>- CuedColor: a number identifying the cued color [1..8]</p><p>- Subject: a number identifying the participant in experiment 1 [1..14]</p><p><br></p><p>* exp1_ResponseTimes.csv</p><p>(Figure 4, time between array presentation and key-press response)</p><p><br></p><p>- ResponseTimes: the response time in seconds</p><p>- Condition: 1=positive cue, 2=negative cue</p><p>- ArraySize: number of items in the search display [6,12,18,24 or 30]</p><p>- TargetPresence: 1=target was present, 2=target was absent</p><p>- Subject: a number identifying the participant in experiment 1 [1..14]</p><p><br></p><p>* exp1_proportionFixationsRelevant.csv</p><p>(Figure 5 top, proportion of fixations landing on search-relevant items, for target absent arrays only)</p><p><br></p><p>- Proportion: the proportion of fixations landing on items that could contain the target, given the cue</p><p>- CueType: 1=positive cue, 2=negative cue</p><p>- ArraySize: number of items in the search display [6,12,18,24 or 30]</p><p>- Subject: a number identifying the participant in experiment 1 [1..14]</p><p><br></p><p>* exp1_Refixations.csv</p><p>(Figure 5 bottom, refixation rate for target absent arrays only)</p><p><br></p><p>- Refixations: average number of fixations per fixated item</p><p>- CueType: Condition: 1=positive cue, 2=negative cue</p><p>- ArraySize: number of items in the search display [6,12,18,24 or 30]</p><p>- Subject: a number identifying the participant in experiment 1 [1..14]</p><p><br></p><p>* exp1_development.csv</p><p>(Figure 6, development of guidance through first 5 fixations)</p><p><br></p><p>- proportion_relevant: proportion of fixations on search-relevant items</p><p>- subject: a number identifying the participant in experiment 1 [1..14]</p><p>- fixation: the fixation within the search [1..5] (fixation 0 triggers array onset and is not used here)</p><p>- cue_type: -1=negative cue, 1=positive cue</p><p><br></p><p><br></p><p>Files for experiment 2:</p><p>-----------------------</p><p><br></p><p>* exp2_performance_RT.csv</p><p>(Figure 7, response times, as well as the not plotted correctness of key-press responses)</p><p><br></p><p>- subject: a number identifying the participant in experiment 2 [1..14]</p><p>- cue_type: -1=negative cue, 0=neutral cue, 1=positive cue</p><p>- no_rel_colors: number of relevant colors (identifying items that could contain the target)</p><p>- array_size: number of items in the search display [12 or 24]</p><p>- target_presence: 0=target not present, 1=target present</p><p>- asnwer: 0=response not correct, 1=response correct</p><p>- RT: response time in miliseconds</p><p><br></p><p>* exp2_proportion_on_relevant.csv</p><p>(Figure 8, the proportion of fixations landing on search-relevant items, for target absent arrays only)</p><p><br></p><p>- subject: a number identifying the participant in experiment 2 [1..14]</p><p>- arraysize: number of items in the search display [12 or 24]</p><p>- cuetype: -1=negative cue, 0=neutral cue, 1=positive cue</p><p>- n_rel_cols: number of relevant colors (identifying items that could contain the target)</p><p>- proportion_relevant: the proportion of fixations landing on items that could contain the target, given the cue and array</p><p><br></p><p>* exp2_development</p><p>(Figure 9, development of guidance by the cues in the first 5 fixations)</p><p><br></p><p>- subject: a number identifying the participant in experiment 2 [1..14]</p><p>- cuetype: -1=negative cue, 0=neutral cue, 1=positive cue</p><p>- arraysize: number of items in the search display [12 or 24]</p><p>- n_rel_cols: number of relevant colors (identifying items that could contain the target)</p><p>- fix1: proportion of first fixations on search-relevant items</p><p>- fix2: proportion of second fixations on search-relevant items</p><p>- fix3: proportion of third fixations on search-relevant items</p><p>- fix4: proportion of fourth fixations on search-relevant items</p><p>- fix5: proportion of fifth fixations on search-relevant items</p>