recognition

• Visual word recognition

  • Potential parallel input - multiple words often visible = potential processing several words at once
  • Orthography - visual representation
  • Clear lexical segmentation
  • Learned ability - Becomes automatic
• Auditory word recognition

  • Serial input - sequentially, 1 at a time
  • phonetics/ phonology - sounds & how sounds organized
  • No clear segmentation cues
  • Unaided acquisition - naturally through exposure
• Character recognition - template matching 


  • storing templates or patterns of characters in brain of every possible input will look like
  • match observed objects to proper image in memory
  • storage concerns - all possible representations of letters?
  • normalisation to letter prototypes - cleaning up images, dealing with occlusion & noise
• Character recognition - feature detection 


  Analysis by synthesis

  1. character broken down to constituent parts
  2. list of parts compared to patterns in memory
  3. best matching pattern chosen
• Word superiority effect (Reicher 1969)


  • letters recognized more accurately & rapidly when presented in meaningful context = Lexical context = evidence of top-down control of character recognition
  • bias towards 'well formed' stimuli - misidentify words with uncommon spelling & non-words as neighbouring words
• Interactive activation model - McClelland & Rumelhart 1981


  integrates bottom-up & top-down information processing - provides word-level context effects
• Interactive activation model - McClelland & Rumelhart 1981 - Similar to TRACE


  • incorporates (visual) features
  • (positional) characters - lateral inhibition (activation of 1 unit inhibits activation of neighboring), excitatory & inhibitory connections between units
  • word detectors - lateral inhibition
• Routes to visual word recognition - role of phonology -

  Assembled route hypothesis 


  • Visual analysis of characters
  • = Activation of phoneme corresponding to character
  • activation = helps remaining lexical access as Spoken Word Recognition - phonological information helps brain recognize & understand word by linking visual input to spoken form
• Routes to visual word recognition - orthography

  • system of writing in a language - rules, conventions for spelling, punctuation, grammar
  • varies across languages & writing systems
• Routes to visual word recognition - shallow orthography

  • more consistent & predictable spelling rules
  • 1 relationship between letters & phonemes
  • = direct & consistent relationship between letters used in writing & phonemes they represent in speech
  • Each letter typically corresponds to 1 specific sound = easier to predict pronunciation based on spelling
  • Spanish, Finnish
• Routes to visual word recognition - deep orthography

  • more complex & variable spelling patterns
  • Same letter often used to signify different sounds in different contexts
  • Some advantage: morphologically related but different sounding words spelled similarly
  • = related in meaning but different sounds may be spelt similarly
  • = reflects morphological connections between words
  • Hebrew (heavy), English (less so)