Table 2.

Model ROC area for face recognition memory. Image likelihoods are determined by combining the familiarities of image fragments using either naive Bayes (Equation 7) or the mean of the fragment familiarities (Equation 8). The likelihood of an image given the distracter class is found using a background model with either 10 or 80 dimensions. Standard errors of the mean are computed over 5 random trials.

		ROC area
Kernel	Fragment combination	10-D BG	80-D BG
Gaussian (σ = 1)	Naive Bayes	0.94 t 0.03	0.58 ± 0.02
	Mean familiarity	0.97 ± 0.02	0.62 ±0.13
kNN (k = 1)	Naive Bayes	0.93 ± 0.05	0.97 ± 0.02
	Mean familiarity	0.96 t 0.02	0.96 ±0.01