Information Processing Center, Osaka Kyoiku University Kashihara, Japan.
Department of Human Sciences, Osaka Prefecture University Sakai, Japan.
Front Psychol. 2015 Feb 16;6:76. doi: 10.3389/fpsyg.2015.00076. eCollection 2015.
It is widely known that statistical learning of visual symbol sequences occurs implicitly (Kim et al., 2009). In this study, we examined whether people can learn the serial order of visual symbols when they cannot detect them. During the familiarization phase, triplets or quadruplets of novel symbols were presented to one eye under continuous flash suppression (CFS). Perception of the symbols was completely suppressed by the flash patterns presented to the other eye [binocular rivalry (BR)]. During the test phase, the detection latency was faster for symbols located later in the triplets or quadruplets. These results indicate that serial order learning occurs even when the participants cannot detect the stimuli. We also found that detection became slower for the last item of the triplets or quadruplets. This phenomenon occurred only when the participants were familiarized with the symbols under CFS, suggesting that the subsequent symbols interfered with the processing of the target symbol when conscious perception was suppressed. We further examined the nature of the interference and found that it occurred only when the subsequent symbol was not fixed. This result suggests that serial order learning under BR is restricted to fixed order sequences. Statistical learning of the symbols' transition probability might not occur when the participants cannot detect the symbols. We confirmed this hypothesis by conducting another experiment wherein the transition probability of the symbol sequence was manipulated.
众所周知,视觉符号序列的统计学习是隐式发生的(Kim 等人,2009)。在这项研究中,我们考察了当人们无法察觉到视觉符号时,他们是否能够学习这些符号的序列顺序。在熟悉阶段,三胞胎或四胞胎的新符号在连续闪光抑制(CFS)下呈现给一只眼睛。通过向另一只眼睛呈现闪光图案来完全抑制对符号的感知[双眼竞争(BR)]。在测试阶段,位于三胞胎或四胞胎中较后的符号的检测潜伏期更快。这些结果表明,即使参与者无法检测到刺激,序列顺序学习也会发生。我们还发现,三胞胎或四胞胎的最后一项的检测速度会变慢。只有当参与者在 CFS 下熟悉符号时才会出现这种现象,这表明当有意识的感知被抑制时,后续符号会干扰目标符号的处理。我们进一步研究了干扰的性质,发现只有当后续符号不固定时才会发生干扰。这一结果表明,在 BR 下的序列顺序学习仅限于固定顺序序列。当参与者无法检测到符号时,符号的转移概率可能不会发生统计学习。我们通过进行另一个实验证实了这一假设,该实验中操纵了符号序列的转移概率。