The current study examined the influence of eyewitness familiarity with the defendant (familiar vs. not familiar), type of descriptor discrepancy (none vs. permanent vs. non-permanent), and eyewitness age (10- vs. 20-year-old) on mock jurors’ decision making. Mock jurors (N = 422 undergraduate students) read a mock trial involving an alleged motor theft. Familiarity with the defendant resulted in more guilty verdicts and higher guilt ratings compared to when the relationship between the eyewitness and defendant was described as ‘strangers’. Mock jurors also were more likely to reach a guilty verdict, provide higher guilt ratings, and rate the eyewitness more favourably when no discrepancies were reported compared to when there were discrepancies. Moreover, the defendant was rated less favourably when discrepancies were present. Surprisingly, mock jurors were more likely to vote guilty, and rate the eyewitness more favourably and the defendant less favourably, when descriptor discrepancies were permanent features made compared to non-permanent feature discrepancies. No effects of eyewitness age were found.