The handle of literacy: evidence from preliterate children and illiterate adults on orientation discrimination of graspable and non-graspable objects

ABSTRACT

Prior research reported an apparently developmental trajectory on mirror-image discrimination (Gregory, Landau, & McCloskey, 2011): reflections across the object principal axis (OPA) and across the external vertical axis (EVA) are hard to discriminate by children, but only OPA reflections are problematic for adults. In this study, we investigated how literacy acquisition and object visuomotor properties affect this trajectory. Six-year-old children (preliterate preschoolers vs. first graders: Experiment 1) and illiterate, ex-illiterate, and schooled literate adults (Experiment 2), searched for graspable (e.g. hammer) or non-graspable (e.g. sock) target-objects amongst orientation-contrast distractors. OPA and EVA errors were predominant in non-readers, but EVA errors dropped sharply in readers. Graspability enhanced OPA and EVA mirror-image discrimination, especially in non-readers. Thus, the reduction of EVA mirror-image errors is not driven by maturation, cognitive development, or schooling; the underlying mechanisms are mostly learning to read and to a smaller extent the operation of the dorsal stream.

KEYWORDS:

• Literacy acquisition
• dorsal stream
• orientation
• mirror-image discrimination
• visuomotor affordances

Acknowledgements

We express our gratitude to Emma Gregory for sharing with us the experimental material of her studies. We thank Eugénia Madureira and Maria Sofia Pinheiro for the assistance in experimental testing and Luísa Corbal for her help on the analysis run on pixel overlap. We are also very grateful to Régine Kolinsky, one anonymous reviewer, and Jon A. Duñabeitia for their comments on the prior version of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Tânia Fernandes http://orcid.org/0000-0002-4453-7251

Notes

1 The locus of mirror discrimination during visual recognition has been debated (see, e.g. Dehaene, Nakamura, et al., 2010; Duñabeitia, Molinaro, & Carreiras, 2011; Pegado, Comerlato, Ventura, et al., 2014; Perea, Moret-Tatay, & Panadero, 2011) and some studies have exclusively examined orthographic stimuli (letters and words; e.g. Duñabeitia et al., 2011; Perea et al., 2011). Given that the neurocognitive mechanisms responsible for mirror discrimination of nonlinguistic objects vs. orthographic stimuli do differ (e.g. Davidoff & Warrington, 2001; Pegado et al., 2011; Priftis et al., 2003), a thorough revision of the studies on orthographic stimuli is out of the scope of the present study.

2 The 3DM reading fluency test comprises three lists (high-frequency words, low-frequency words, and nonwords); participants are asked to read aloud as many items as possible in 30 s per list. Performance was computed as the total of items correctly read across list (in 3×30 s).

3 The LL3 is a paper-and-pencil, multiple-choice, sentence-completion test that requires silent reading; participants are asked to select by circling the word (out of five) that adequately completes each sentence in 5 min for as many sentences as possible (out of 36).

4 The objects identified with * were selected for this study and are presented in the supplemental material. The other objects can be found in Gregory et al. (2011).

5 We thank an anonymous reviewer for this suggestion.

6 In each trial, participants had to choose the target out of 8 stimuli; thus, a performance at chance corresponds to correct responses in 1/8 of trials, i.e. 12.5% correct responses.

7 The errors of each orientation-contrast are not orthogonal to each other, which violates the sphericity assumption of ANOVA. Therefore, the Greenhouse–Geisser correction was applied. The corrected p-values are reported, but with the original degrees of freedom.

8 The 90° condition corresponds to the mean number of errors across −90° and +90° plane-rotation contrasts given that we had no reason to expect differences between them.

Additional information

Funding

This work was supported by IF 2013 program of the Fundação para a Ciência e a Tecnologia, FCT, Portugal, under grant number IF/00886/2013/CP1194/CT0002, by the Research Center for Psychological Science at Universidade de Lisboa (CICPSI), and by grant number UID/PSI/00050/2013 and grant number POCI-01-0145-FEDER-007294 from FCT (COMPETE and FEDER programs) to the Center for Psychology at the University of Porto.