Distinct Reading Routes For Deep and Shallow Orthographies in Simultaneous Biscriptals- A Functional Imaging Study Tanusree 1National

1 Das ,

Uttam

Brain Research Centre, Manesar, India/

1 Kumar ,

2University

Raju S

2 Bapi ,

Prakash

3 Padakannaya ,

of Hyderabad, Hyderabad,

Nandini C

3 India/ University

1 Singh (tanusree, [email protected])

of Mysore, Mysore, India

317SA-AM

Results: Behaviour Experiment 3: Word reading in Hindi monoscriptals

Introduction

High>Low

The process of reading can be achieved through two distinct routes, the indirect (dorsal) route and the direct (ventral) route. Reading studies with monoscriptals have established cultural differences (1) wherein monoscriptal readers of a shallow orthography like Italian follow the indirect route for reading, whereas monoscriptal readers of a deep orthography like English follow both the direct and the indirect routes. Studies with biscriptals are limited. Late biscriptals have shown that neural mechanisms of second-learnt orthography are shaped by the first (1, 2). Our objective is to investigate neural mechanisms for reading in early biscriptals, who simultaneously learn to read both deep and shallow orthographies.

L-IPL Low>High (dorsal route)

Hypothesis For early Hindi-English biscriptals, we hypothesize orthography-specific routes to reading. Thus, for Hindi which is shallow, we hypothesize the use of indirect reading route wherein graphemes are assembled sequentially to form a word. For English, which is deep, we postulate the recruitment of both the direct and indirect reading routes. Previous studies have shown that direct route is a ventral route that goes through:

Reading both high and low frequency words only showed activation along the dorsal route (IPL). Experiment 4: Word reading in English monoscriptals High>Low

Occipital lobe---------->left ITG/MTG---------->left IFG (3)

Low>High (dorsal and ventral route)

L-IPL

L-ITG

and the indirect route is a dorsal route that goes through: Occipital lobe---------->left IPL---------->left IFG (4)

Participants Early biscriptals 10 right-handed Hindi-English biscriptals in the age group 25-32 years volunteered for the study. A separate group of Hindi monoscriptals and English monoscriptals in the same age group were recruited as controls. All the participants were assessed for reading proficiency based on the tasks outlined below.

Early biscriptals were equally fluent in reading English and Hindi. Reading both high and low frequency words only showed activation along the dorsal (IPL) and ventral route (ITG).

Assessment for reading proficiency Questionnaire 1. Self assessment: Participants were asked to rate proficiency in conversation, reading and writing in Hindi and English on a scale of 0-10. Behaviour 1. Reading time: Participants read passages in English and Hindi and time taken to read the passage was measured as a test for reading fluency. 2. Response time: Participants read aloud a set of 120 high frequency and 120 low frequency nouns in English and Hindi, outside the scanner and were recorded. The nouns were matched for number of syllables in each orthography . Time taken to respond to a word and time taken to read the words were calculated using a Matlab software. Spectral analysis 1. Articulation space: Participants were recorded during the passage reading task. Their speech was analysed for different articulatory features using novel spectral analysis techniques. The energy distribution of these articulatory features are represented as an articulatory space also called the speech modulation spectrum (5). The area occupied by the articulation space provides an objective measure of reading fluency.

Results: BOLD analysis

In early biscriptals, significant differences in reaction time and reading time were seen between high and low frequency words in English but not in Hindi. This finding was also observed in English monoscriptals and could indicate a difference in reading strategy adopted for high and low frequency words in English but not in Hindi. Based on our behavioral findings we postulate differences in reading strategies for English and Hindi respectively.

Methods: fMRI Experiment 1: Phrase reading by early biscriptals

Results: fMRI

Conclusions

Experiment 1: Phrase reading in Early biscriptals Early English

1) High frequency words in both Hindi and English follow the ventral (MTG/ITG) route.

Early Hindi

Early biscriptals overtly read short phrases during the block design paradigm in a Phillips 3T MRI Scanner. The phrases were extracted from a passage ‘North wind and the Sun’ and its Hindi translation called ‘Uttari hawa aur suraj’. They were semantically matched in both orthographies. Two runs/subject were administered. Whole-brain structural T1-weighted images and functional T2-weighted images were obtained with the following parameters: TR/TE=2s/35ms, flip angle =90 degrees, FOV =230 mm with 64X64 image matrix, with an in plane resolution of 3.59X3.59 mm. Slice thickness= 4mm with 1 mm gap. Total volumes/subject= 320. The acquired images were analysed in SPM 5. A word reading paradigm consisting of high and low frequency words were administered in experiments 2, 3 and 4.

2) The difference in reading strategies is manifested while reading low frequency words. The early biscriptals follow both dorsal (IPL) and ventral (ITG) routes while reading low frequency words in English. For reading low frequency words in Hindi, they predominantly use the dorsal route (IPL). The cortical activations showed distinct pathways for reading English and Hindi. For English, the dorsal and the ventral route are recruited for reading phrases. For Hindi, in addition to the ventral route (ITG), we also see a route through left MTG.

References

Experiment 2: Word reading by early biscriptals

Experiment 2: Word reading in Early biscriptals Reading route in English

Early biscriptals overtly read 240 words in English and Hindi. Words were concrete nouns which belonged to two categories, namely high frequency words and low frequency. For each orthography, 120 words were presented of which 60 were high-frequency and 60 in low-frequency. Each run consisted of 12 blocks, 6 blocks of reading interspersed with 6 blocks of rest. Of the 6 blocks, 3 blocks used high frequency words and 3 used low frequency words. 10 words were presented in each block that lasted 20 seconds. Hindi words chosen were all linear concrete nouns and no nonlinear words with vowels (or matras) around the consonants were used. For English, no irrregular (yacht) words were used. This ensured that the non-linearity of Hindi was avoided but the alphabetic nature was preserved and matched with English. The orthographies now only differed in orthographic structure - Hindi was shallow, and English was deep. One experimental run time= 240 s TR= 2 s, total number of volumes/subject= 480. Remaining parameters and analysis protocol were similar to experiment 1.

High>Low (ventral route)

Reading route in Hindi L-MTG

3) Jobbard, G., Crivello, F. & Mazoyer-Tzourio, N. (2003), ‘Evaluation of the dual route theory of reading: A metanlysis of 35 neuroimaging studies’, Neuroimage, vol. 20, pp. 693-712. L-MTG 4) Nakamura, K., Hara, N., Kouider, S., Takayama, Y., Hanajima, R., Sakai, K & Ugawa, Y. (2006), ‘Task-giuded selection of the dual neural pathways for reading’, Neuron, vol. 52, 557-564. 5) Singh, L. & Singh, N.C. (2008), ‘The development of articulatory signatures in children’, Developmental Science, vol. 11 (4), 467-473. High frequency words (English): ventral route (left MTG) High frequency words (Hindi): dorsal (left IPL) and ventral (left MTG) route L-IPL Low>High (dorsal route)

Low>High (dorsal and ventral route)

L-IPL Acknowlegements We acknowledge NBRC for supporting this research, Prof P. Raghunathan, Mr. Jitender and Ms Sumathi for fMRI acquisition; Ms Megha Sharda for help with data analysis and Ms Padma Subhadra for help with poster designing. We acknowledge our participants.

L-ITG

Experiment 4: Word reading by English monoscriptals The word reading task in English was administered to English monoscriptals selected using the above-mentioned criteria.

2) Kumar, U. & Das, T., Bapi, R.S., Padakannaya, P., Joshi, R.M. & Singh, N.C., (2009), ‘Reading different orthographies: an fMRI study of phraase reading in Hindi-English bilinguals’, Reading and Writing, doi: 10.1007/s11145-009-9176-8.

High>Low (dorsal and ventral route)

Experiment 3: Word reading by Hindi monoscriptals The word reading task in Hindi was administered to Hindi monoscriptals selected using the above-mentioned criteria.

L-IPL

1) Paulesu, E., McCrory, E., Fazio, F., Menoncello, L., Brunswick, N., Cappa, S. F., Cotelli, M., Cossu, G., Corte, F., Lorusso, M., Presenti, S., Gallagher, A., Perani, D., Frith, C. D. & Frith, U. (2000), ‘A cultural effect on brain function’, Nature Neuroscience, vol. 3, no. 1, pp. 91-96.

Low frequency words (English): dorsal (left IPL) and ventral (left ITG) route Low frequency words (Hindi): dorsal (left IPL) route