Background. When accidents occurred, eyewitnesses become very important for the police to get the clear description of the case. Unfortunately, eyewitness testimony is easily corrupted with error called false memory. Previous studies showed that people that been trough traumatic events tend to falsely recalled DRM list related to their trauma. However, false memory studies about traffic accident have not been found. This study hypothesizes that people that had been trough traffic accident will falsely recalled trauma related words (accident related) than neutral and negative lists. Besides word types as external factors, this study also examined gender as internal factors. Previous studies still found contradictory results related the effect of gender as internal factors on false memory. The effect of interaction between word types and gender also producing contradictory results. The aim of this study was to examined the effect of word types and gender on the formation of false memory.

Methods. Participants were 60 motorcycles driver (M_age = 19.9, 30 females) that at least had been through an accident while driving motorcycles. These participants were recruited from online form. This study is experimental mixed design factorial ANOVA 3 (word types) x 2 (gender) with word types as within-subject factor and gender as between-subject factor. False memory was measured by recall and recognition test. Participants were given nine DRM word list, consisted of 3 neutral, 3 negative and 3 accident-related lists. The sequence of word list was counterbalanced. Participants were presented each list that consisted of ten words. After each list was presented, participants were given ditraction task and then recall test. The same procedure was repeated until the ninth lists. Then, in recognition test participants were asked to make Remember-Know judgement if they recognize the word.

Results. Based on recall test, there was main effect of word types, F(2, 164) = 4.83, p = .01, ƞ²= .077. Accident lists (M = .48, SD = .62) were significantly more falsely recalled than neutral (M = .28, SD = .52) and negative lists (M = .22, SD = .41). No significant effect of gender, F(1, 82) = .39, p > .05, ƞ²= .007, nor interaction of word types and gender, F(2, 164) = 2.59, p > .05, ƞ²= .043, was found. The result on recognition test showed a significant main effect of word type, F(2, 164) = 5.13, p = .007, ƞ²= .081, where accident lists (M = 1.80, SD = .95) were more falsely recognised than neutral (M = .1.45, SD = .87) and negative lists (M = 1.48, SD = .85). However, there were no main effect of gender, F(1, 82) = 1.77, p > .05, ƞ²= .03, nor the interaction effect, F(2, 164) = .24, p > .05, ƞ²= .004.

Conclusions. Word types has effect on formation of false memory, more specifically accident list is more falsely recalled and recognized more than neutral and negative list in recall and recognition test. This study didn’t find any effect of gender nor interaction of word types and gender towards false memory in any of the tests.

Added-Values. This study is the start of false memory studies on traffic accident context. Authorities must be more cautious when asking for eyewitness testimonies, especially those that include traffic accident words because those words may lead to false memory. When questioning, it is more recommended to use neutral words to reduce false memory in the testimony.

Keywords: false memory, word types, gender, traffic accident, motorcycle drivers

Word counts: 561