Data fields for files 'DataX1' and 'DataX2'
Tab-delimited .txt

sub = arbitrary numeral encoding subject identity
cond = phonotactic condition as described in paper: 'anti' = anti-spirantization; 'len1' = spirantization 1; 'len2' = spirantization 2; 'AntiV' = anti-voicing; 'LenV' = voicing
trial = ordinal from 0 to 29 indicating serial position of trial in experiment
resp = raw keyboard response: 'a' = 'first word'; 'l' = 'second word', 'None' = timeout
rt = reaction time in ms
word1 = arbitrary numeral encoding first word's identity
word2 = arbitrary numeral encoding second word's identity
targfirst = logical indicating whether the first word was the target or not
respfirst = logical indicating whether subject responded 'first word'; timeouts are left blank
correct = logical indicating whether response was correct; timeouts are assigned 0
Language variables: for experiment1, where there were a lot of bilinguals, these are a series of dummy variables indicating fluent English, Mandarin ('chi'), Spanish, and other. For experiment2, which had far fewer bilinguals, this is simply coded as a text response 'lang'.
word1phon = rough transcription of word 1. Note that all consonants are transcribed as voiced stops. Phonetic continuancy for experiment 1 and phonetic voicing for experiment 2 are predictable from the condition and which word is the target. For instance, any target in the spirantization condition of experiment 1 will have an initial stop and medial continuants. 
word2phon = rough transcription of word 2. Note that all consonants are transcribed as voiced stops. Phonetic continuancy for experiment 1 and phonetic voicing for experiment 2 are predictable from the condition and which word is the target. For instance, any target in the spirantization condition of experiment 1 will have an initial stop and medial continuants.