Areas of My Research
Healthy Control Readers
Individual Differences in Reading
Readers with Schizophrenia
Dias et al. (2021)
Individuals with schizophrenia show dramatic reading fluency deficits (Vanova et al., 2020), which are attributed to dysfunction in both lower-level (e.g., oculomotor control, parafoveal) and higher-level (e.g., lexical access, semantic integration) processes according to the two-hit deficit model (Dias et al., 2021).
Much of the work examining eye-movements during reading in individuals with schizophrenia utilizes single-word or single-line reading paradigms (Vanova et al., 2020), and therefore the results do not inform on eye-movements unique to connected-text reading, such as return-sweep saccades (i.e., eye-movement made from the end of a line to the beginning of the next line).
Consistent with the two-hit deficit model, readers with schizophrenia show evidence of lower-level, processing deficits when reading multi-line passages, such that readers with schizophrenia make more return-sweep saccade targeting errors and show reduced sensitivity to upcoming parafoveal information compared to healthy controls (Christofalos et al., 2023).
Disruptions to Early Word Recognition During Reading
Inter-word spaces provide information on where to move our eyes next during reading (Li et al., 2009).
Removing spaces (i.e., word boundary cues) delays early lower-level word identification processes during single-sentence reading (Sheridan et al., 2016). However, it's unclear how removing inter-word spacing impacts higher-level processes during connected-text reading.
My current eye-tracking work examines how delays in early word identification (by removing spaces) impacts later inference processing during connected-text reading to provide insight into the interaction between lower- and higher-level processes during reading.
Li et al. (2009)
Emoji Semantically Congruent with Passage Inference
Emoji Semantically Congruent with Target Word in Passage
Emoji Processing During Reading
Emojis facilitate reading comprehension and are processed similar to words (Barach et al., 2021).
When reading texts containing emojis, the eyes move quickly across the text to process the emoji meaning (Howman & Filik, 2020). My current eye-tracking work examines how this earlier access of semantic information portrayed by emojis facilitates higher-level (i.e.., inference) processing.
Emojis are visually-compact, which means that upcoming semantic information conveyed in emojis may be pre-processed in the parafovea prior to fixation. Another area of my current eye-tracking work examines parafoveal processing of emojis during reading.
Impact of Titles on Situation Model Comprehension
Spending more time looking at chapter headings during reading is associated with higher reading comprehension scores (de Leeuw et al., 2016).
Titles and headings improve comprehension by serving as cues or advanced organizers that provide structure for interpreting upcoming information (Dooling & Mullet, 1973).
Readers show a large reduction in situation model comprehension when reading difficult, vague passages with no title compared to when a title is present, suggesting that titles provide a theme that helps readers to develop a coherent situation model representation when reading (Christofalos et al., 2020).
The effects of title presence are much smaller on both surface form and textbase comprehension, suggesting that titles uniquely impact situation model comprehension (Christofalos et al., 2020).
Impact of Attention on Situation Model Comprehension
Shifting and updating one's situation model representation during reading requires significant executive function resources (Butterfuss & Kendeou, 2018).
Resource Depletion Theory suggests that executive control is a limited cognitive resource that is depleted over sustained periods of mental exertion (Muraven & Baumeister, 2000).
A depletion of attentional resources (as a result of performing a difficult typing task) significantly reduces readers' ability to construct a coherent situation model representation when reading difficult, vague passages. Interestingly, resource depletion did not significantly impact other levels of comprehension (i.e., surface form and textbase), suggesting that constructing a coherent situation model uniquely depends on the availability of sufficient attentional resources (Christofalos et al., 2022).
Neural Substrates of Reading
During visual processing, the left hemisphere specializes in processing information on a local (i.e., detail) level , whereas the right hemisphere specializes in processing information on a global (i.e., big picture) level (Stiles et al., 2008). Similar processes occur during connected-text reading (Beeman et al., 1998).
The right hemisphere's overlapping and diffuse semantic fields allow for efficient integration across multiple word meanings (i.e., semantic integration).
Efficient right hemisphere semantic integration processes are important when updating one's situation model representation during reading, as evidenced by significant increases in activation in right anterior temporal lobe when readers update their situation model representation (Branzi et al., 2020).
Right Anterior Temporal Lobe
CRA Problem Salivi et al. (2020)