The Significance of Letter Position in Word Recognition

By Graham Rawlinson. 2007

115 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Cognitive Science & Psychology, Linguistics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yonatan Belinkov
2. James R. Glass
3. Fahim Dalvi
4. Nadir Durrani
5. Hassan Sajjad
6. Mohit Bansal
7. Percy Liang
8. Samuel R. Bowman

In Computer Vision and Pattern Recognition

1. Yonatan Belinkov
2. Fahim Dalvi
3. Hassan Sajjad
4. Nadir Durrani
5. Preslav Nakov
6. Yonatan Bisk
7. James R. Glass
8. Zachary Chase Lipton

In Cognitive Science & Psychology

1. Ioannis A. Kakadiaris
2. Christophoros Nikou
3. Michalis Vrigkas
4. Yonatan Belinkov
5. Zachary Chase Lipton
6. Yonatan Bisk
7. Bhuwan Dhingra
8. Danish Pruthi

In Linguistics

1. Yonatan Belinkov
2. Zachary Chase Lipton
3. Yonatan Bisk
4. Bhuwan Dhingra
5. Danish Pruthi

Citations in the corpus

• Synthetic and Natural Noise Both Break Neural Machine Translation (2017, background): “Humans have surprisingly robust language processing systems which can easily overcome typos, misspellings, and the complete omission of letters when reading (Rawlinson, 1976).”
• HotFlip: White-Box Adversarial Examples for Text Classification (2017, background): “After a few character changes, the meaning of the text is very likely to be preserved or inferred by the reader (Rawlinson, 1976), which was also confirmed by our human subjects study.”
• Theories of reading should predict reading speed (2012, unknown): “Rawlinson provided a copy of his thesis and granted permission to post it on the Web (Rawlinson 1976).”
• Combating Adversarial Misspellings with Robust Word Recognition (2019, methodology): “ScRNN Model Inspired by the psycholinguistic studies (Matt Davis, 2003; Rawlinson, 1976), Sakaguchi et al.”
• A Review of Human Activity Recognition Methods (2015, background): “The intuition behind this approach is a psycholinguistics phenomenon, where randomizing letters in the middle of words has almost no effect on understanding the underlying word if and only if the first and the last letters of this word remain unchanged (Rawlinson, 2007).”
• How Robust Are Character-Based Word Embeddings in Tagging and MT Against Wrod Scramlbing or Randdm Nouse? (2017, background): “Word scrambling is motivated from psycholinguistic studies (Rawlinson, 1976).”
• One Size Does Not Fit All: Comparing NMT Representations of Different Granularities (2019, background): “In Middle errors, the order of the first and the last characters of a word are preserved, while the middle characters are randomly shuffled (Rawlinson, 1976), e.”