Lange Gesa, De Weerd Peter
Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands.
Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, Maastricht, The Netherlands; Maastricht Centre for Systems Biology (MaCSBio), Maastricht University, Maastricht, The Netherlands.
Vision Res. 2018 Feb;143:103-116. doi: 10.1016/j.visres.2017.06.018. Epub 2017 Dec 8.
Substantial transfer of perceptual skill learning can be achieved across large distances in the visual field by a brief pre-test, training-plus-exposure, or a double-training paradigm (Xiao et al., 2008; Zhang, Xiao, et al., 2010; Zhang, Zhang, et al., 2010). Additionally, subliminal exposure has been shown to be beneficial for subsequent perceptual learning. Here, we tested the generalization of orientation discrimination learning from a fully trained location towards four other test locations, either in the same or opposite hemifield as the training location, which each were subjected to a different type of pre-conditioning. In one test location, there was brief pre-testing in the first session. Two other locations were stimulated by masked stimuli similar or identical to concurrently presented stimuli in the training location. In the fourth test location, no stimuli were presented during training. Generalization of training to test locations was measured in the session immediately following the completion of training in the training location. Moreover, to test the robustness of transfer, training was continued in all four test locations. The experiment as a whole consisted of 15 sessions of orientation discrimination learning at the training location, followed by 15 sessions of training in the test locations. We found only limited generalization from the trained to the test locations. Performance in pre-tested and stimulated test locations showed a small advantage compared to the unstimulated test location. However, this advantage disappeared within a few sessions of further training in the test locations.
通过简短的预测试、训练加暴露或双重训练范式，可在视野中远距离实现感知技能学习的实质性迁移（Xiao等人，2008年；Zhang、Xiao等人，2010年；Zhang、Zhang等人，2010年）。此外，阈下暴露已被证明对后续的感知学习有益。在此，我们测试了从一个完全训练的位置向其他四个测试位置的方向辨别学习的泛化情况，这些测试位置与训练位置在同一或对侧半视野，每个测试位置都接受了不同类型的预处理。在一个测试位置，在第一阶段进行了简短的预测试。另外两个位置受到与训练位置同时呈现的刺激相似或相同的掩蔽刺激的刺激。在第四个测试位置，训练期间不呈现刺激。在训练位置完成训练后的紧接着的阶段中测量训练到测试位置的泛化情况。此外，为了测试迁移的稳健性，在所有四个测试位置继续进行训练。整个实验包括在训练位置进行15个阶段的方向辨别学习，随后在测试位置进行15个阶段的训练。我们发现从训练位置到测试位置的泛化有限。与未受刺激的测试位置相比，经过预测试和受刺激的测试位置的表现有小的优势。然而，在测试位置进一步训练几个阶段后，这种优势就消失了。
