Suzy Styles | Nanyang Technological University (original) (raw)

Uploads

Papers by Suzy Styles

The TOWRE-2 Test of reading is well recognized in educational research in many English-speaking c... more The TOWRE-2 Test of reading is well recognized in educational research in many English-speaking counties. Deciding what is the 'correct' pronunciation may be complex in some varieties of English. Audio recordings can a) facilitate creation of a reference repository which may help future coders with their decisions; b) provide training data for automated scoring support. The Spy School Wrapper provides a narrative frame to make wearing a head-mounted microphone more fun for children of primary school age, and places the reading task in the context of a fun game.

Poster presented at the International Multisensory Research Forum, Amsterdam, June 2014 ... more Poster presented at the International Multisensory Research Forum, Amsterdam, June 2014 Abstract. 78. What can ancient scripts tell us about sound symbolism? Suzy Styles Division of Psychology, Nanyang Technological University Do English-speakers encode sound/shape symbolism for ‘i’ and ‘u’ because of the shape of the letter-forms? Or do languages adopt writing systems which encode salient patterns of sound/shape symbolism? This paper looks to writing systems from across human history to disentangle the direction of causation, by asking whether writing systems visually communicate salient phonological contrasts like /i/ vs /u/. Stimuli were pairs of written letters representing the phonemes /i/ and /u/ in 56 different writing systems from across the globe, throughout human history. All scripts from ‘The World’s Writing Systems’ (Daniels & Bright, eds, 1996) were included if it was possible to represent /i/ and /u/ in isolation. Exclusions were duplicate letter forms and scripts from the Phoenician family (including Greek, Latin, Runic and Cyrillic), with its familiar ‘I’ and ‘U’ shapes. The scripts range from Cuneiform and Mayan hieroglyphs at the earliest, through Devanagari, Ethiopic and Sineform scripts, to Tibetan, Arabic, Hebrew, and modern scripts like Korean Hankul. Results. When asked which letter in each /i//u/ pair was which, participants consistently performed above chance (accuracy: M=60%, SD=6%; p>.001). This finding indicates that across human history, writing systems have been better-than-chance at conveying a highly salient speech contrast. While many of the scripts were not matched systematically, more scripts were consistently matched correctly (9), than incorrectly (3). Analysis of the consistently matched stimuli reveals which visual features of a written letter are most predictive of contemporary sound/shape mappings: complexity (i-simple/u-complex), curvature (i-tight/u-loose), angularity (i-accute/u-obtuse), or vertical position of detail (i-high/u-low). <br

Styles SJ (2014). 'What can ancient scripts tell us about sound symbolism?... more Styles SJ (2014). 'What can ancient scripts tell us about sound symbolism?' Poster presented at IMRF in June 2014, Amsterdam. Summary. Do English-speakers encode sound/shape symbolism for 'i' and 'u' because of the shape of the letter-forms? Or do languages adopt writing systems which encode salient patterns of sound/shape symbolism? This paper looks to writing systems from across human history to disentangle the direction of causation, by asking whether writing systems visually communicate salient phonological contrasts like /i/ vs /u/. Stimuli were pairs of written letters representing the phonemes /i/ and /u/ in 56 different writing systems from across the globe, throughout human history. All scripts from 'The World's Writing Systems' (Daniels & Bright, eds, 1996) were included if it was possible to represent /i/ and /u/ in isolation. Exclusions were duplicate letter forms and scripts from the Phoenician family (including Greek, Latin, Runic and Cyrillic), with its familiar 'I' and 'U' shapes. The scripts range from Cuneiform and Mayan hieroglyphs at the earliest, through Devanagari, Ethiopic and Sineform scripts, to Tibetan, Arabic, Hebrew, and modern scripts like Korean Hankul. Results. When asked which letter in each /i//u/ pair was which, participants consistently performed above chance (accuracy: M=60%, SD=6%; p>.001). This finding indicates that across human history, writing systems have been better-than-chance at conveying a highly salient speech contrast. While many of the scripts were not matched systematically, more scripts were consistently matched correctly (9), than incorrectly (3). Analysis of the consistently matched stimuli reveals which visual features of a written letter are most predictive of contemporary sound/shape mappings: complexity (i-simple/u-complex), curvature (i-tight/u-loose), angularity (i-accute/u-obtuse), or vertical position of detail (i-high/u-low).

Automatic connections between sounds and visual shapes have been documented for some time (c.f., ... more Automatic connections between sounds and visual shapes have been documented for some time (c.f., Spence, 2011). We replicated audiovisual correspondences with simple linguistic sounds /i/ and /u/, this time produced in the lexical tones of Mandarin Chinese, using a modified version of the implicit association test (IAT). Although congruent blocks were significantly faster than incongruent ones (p < .001), no effect of tone congruence was observed. Since tone was an unattended stimulus dimension, we argue that attention modulates sensory congruence in implicit association tasks of this nature.

array. the journal of the ICMA, 2020

A large body of research on connections between sensory modalities has shown that deep connection... more A large body of research on connections between sensory modalities has shown that deep connections exist between sound and vision, such that people have a tendency to associate certain sounds with certain visual properties, including line-drawn shapes. While recognising the role of written language in audio-visual associations, previous research has largely considered written language a potential source of bias rather than a means of gaining deeper in sights into underlying audio-visual associations. We looked to ancient and unfamiliar writing systems spanning recorded human history, to explore whether humans have tried to encode certain characteristics of speech sounds in the letters they created to represent them. Our findings have revealed that modern humans can correctly identify unfamiliar letters at levels higher than would be obtained by chance, and that scripts which encode a particular sound with a particular set of visual characteristics tend to have more correct guesses. ...

array. the journal of the ICMA, 2020

In the world of rock, blues, and popular music, guitarists who pull faces when they play are some... more In the world of rock, blues, and popular music, guitarists who pull faces when they play are somehow more entertaining to watch – as if there is an agreement between the performer and the audience that certain facial gestures enhance the expression of the music at key moments. How does this agreement come about? What is it that viewers and performers share? Research in the field of cross-modal perception has demonstrated that humans share systematic linkages between the senses, including vision and audition. For example, people link higher pitched sounds with smaller, paler, and higher-up objects than lower sounds (e.g., Mondloch & Maurer, 2004). Yet few investigations have explored how these linkages relate to performance in the creative arts. One theory suggests that artists have stronger functional connectivity between sensory areas, and are therefore able to invoke or translate sensory experiences across modalities, using sensory metaphors, and strategic sensory exaggeration (Ra...

"Do infants learn their early words in isolation? Or do they integrate new words into an int... more "Do infants learn their early words in isolation? Or do they integrate new words into an inter-connected semantic system? Between the ages of one and two, infants develop a prodigious word-learning ability (P. Bloom, 2002; Fenson et al., 1994; Hamilton et al., 2000). Little is known, however, about how and when infants integrate their accumulating word-knowledge into an adult-like lexical network – one sensitive to relationships such as association and semantic category organisation. For adults, context has a strong effect on the ease and speed of linguistic processing, and on behavioural responses to language. In lexical priming studies, both visual and auditory context are known to influence the speed of lexical processing (Antos, 1979; Meyer & Schvaneveldt, 1971; Radeau, 1983) and ambiguity resolution (Swinney, 1979), as well as guide predictive looks to contextually relevant areas of a complex scene (Huettig & McQueen, 2007; Kamide et al., 2001). And while priming studies have been conducted with children (Betjemann & Keenan, 2008; Duncan & Kellas, 1983; Hartfield & Conture, 2006; Hashimoto et al., 2007; McCauley et al., 1976; Nation & Snowling, 1999; Radeau, 1983; Schvaneveldt et al., 1977) and toddlers (Krackow & Gordon, 1998), the currently available methods are not applicable below the age of three. This thesis presents a body of research which combines two experimental paradigms into a single new method which allows investigation of relationships within the developing infant lexicon. Over a series of experiments, this thesis develops a version of the infant-friendly Inter-modal Preferential Looking (IPL) task within the lexical priming paradigm. The method combines fine-grained temporal sensitivity with an intuitive, infant-friendly task. Priming effects were observed during lexical access for children two years-of-age and older, providing a promising new direction for investigations into the development of semantic organisation during development."

Materials from ReproducibiliTea sessions in Singapore. Templates and presentations are available ... more Materials from ReproducibiliTea sessions in Singapore. Templates and presentations are available for others to use and edit.

The SEE Cartoons are animated stimuli are designed to attract the eye gaze of infants and toddler... more The SEE Cartoons are animated stimuli are designed to attract the eye gaze of infants and toddlers, for use in developmental research involving onscreen training or testing. The cartoons can be used as engaging inter-trial 'wake-ups', as calming musical interludes, or as rewards for attention. The SEE Cartoons were created by Suzy Styles in 2009 and 2010, for use in Experiments at the Oxford University Babylab. Stills were created in Adobe Illustrator CS2&3, animated in Macromedia/Adobe Flash, with audio edited in Goldwave visual waveform editor. Detailed audio credits and Licensing Information included.

Stimuli, materials and data from experiments investigating sound symbolism for the Tones of Manda... more Stimuli, materials and data from experiments investigating sound symbolism for the Tones of Mandarin Chinese, with speakers of different languages. As reported in Shang & Styles (2017)

ARTEM-IS is a template designed to make reporting EEG/ERP methodology easier and more accurate, b... more ARTEM-IS is a template designed to make reporting EEG/ERP methodology easier and more accurate, by providing specific fields for specific details. We believe the community of expert practitioners can create a more powerful, and more effective tool by working together. This archive includes the ARTEM-IS Statement about the value of a community-agreed tool, a draft template for community consultation, and an archive of documents arising from public events.

A home-sew mask for members of the general public, designed to fit flat at sides, with adjustable... more A home-sew mask for members of the general public, designed to fit flat at sides, with adjustable fitting for a variety of face-shapes. Wearing fabric masks of this type is understood to reduce the transmission of viruses like COVID-19 through droplets. Members of the general public are encouraged to make their own face coverings so that medical grade supplies can be reserved for primary health care workers and first responders. This PDF pattern should be printed at full size (no scaling) on standard A4 paper. V2 corrects sizing and seam allowance. Check latest guidance from the CDC: https://www.cdc.gov/coronavirus/2019-ncov/prevent-getting-sick/diy-cloth-face-coverings.html

array. the journal of the ICMA, 2020

For adults, linguistic processing is automatic and pervasive, arising from the simplest of trigge... more For adults, linguistic processing is automatic and pervasive, arising from the simplest of triggers: viewing a picture of an object whose name is known. Yet little evidence exists about when this process emerges during language learning, and how far the cascade of linguistic activity might progress for children at the beginning of their word learning journey. This talk will present behavioural and electophysiological evidence about how and when picture-viewing generates a cascade of automatic linguistic processing, for infants in the second year of life, and will demonstrate that toddlers' implicitly generated phonological representations are speech-like, accurate, and form part of a co-ordinated system of linguistic processing in which activation of word forms can be linked to the unique characteristics of an individual's growing lexicon.

This paper will present evidence that the toddler lexicon shares substantial similarities with th... more This paper will present evidence that the toddler lexicon shares substantial similarities with the adult lexicon, both in terms of word-to-word lexico-semantic associations, and in terms of phonological competition arising from recent context. Evidence from behavioural paradigms involving eye gaze will demonstrate that the performance of toddlers in comprehension tasks is influenced by semantic context at short (adult-like) priming latencies, and by phonological context arising from adult-like ‘implicit’ labeling of pictures. The time-course of cognitive processes underpinning these effects will be discussed in the light of electrophysiological evidence from ERPs.

Dr Suzy Styles (2012) Supplement created for Language & Cognition, Part 1 Experimental Psychology, Trinity Term 2012.

Some sentences cause a reader to become 'lost' or 'tangled' in an incorrect interpretation of the... more Some sentences cause a reader to become 'lost' or 'tangled' in an incorrect interpretation of the early part of the sentence, meaning that they later have to 're-parse' the sentence, backtracking through the syntax to find a different way of interpreting and coordinating the words. Sentences like these are known as 'Garden Path' sentences. One particularly tricky example is the following...

"The horse raced past the barn fell"

Every year, when I teach this material to Undergraduates at the University of Oxford, about half of the students encountering this sentence for the first time find it very difficult to recover from the garden path, and some can't discover the correct structural interpretation on their own. Here's why it's so difficult, explained according to what we know about online sentence processing, one word at a time.

Supplement created for Language & Cognition, Part 1 Experimental Psychology, Trinity Term 2012. Dr Suzy Styles.

We use an electro-encephalograph (EEG) to visualize the electrical activity generated by our eyes... more We use an electro-encephalograph (EEG) to visualize the electrical activity generated by our eyes, muscles and brain. Participants can either watch a demonstration of EEG recording, or can ‘have a go’ themselves. The EEG is presented on a large screen in front of the participant, where they can see how the electricity in their skin is influenced by moving their eyes, grinning and wiggling their eyebrows. If adult participants are involved, it is sometimes possible to demonstrate alpha-waves as well.

These demonstrations are used to talk about electrochemical processes in the brain, and how EEG can be used to investigate cognitive processes.

The EEG is recorded using four small sensors stuck to the skin of the participant: One sensor is applied to the mastoid bone behind each ear (a reference sensor and a ground). One sensor is applied below the left eye (VEOL), which detects vertical eye movements, and muscle activity from the cheek area. One sensor is applied above the left eye (VEOU), which detects vertical eye movements, and muscle activity from the eyebrows and scalp. The signal from this sensor also includes electrical activity from the brain, as it is positioned over the frontal lobes.

Each sensor contains a small metal electrode plate and a soft sponge, inside a casing at the end of a long, flexible wire. Connection between the skin and the electrode is made by soaking the sponge in a saline solution. Before applying the sensors to the skin with adhesive pads, the skin is cleaned with an alcohol swab, to make sure that normally occurring skin oils do not prevent the electrode from making good contact with the skin. The sensors are connected to a passive EEG analysis system, in which a specialized amplifier sends a boosted signal from each sensor to a PC.

Nonarbitrary mappings between sound and shape (i.e., the bouba-kiki effect) have been shown acros... more Nonarbitrary mappings between sound and shape (i.e., the bouba-kiki effect) have been shown across different cultures and early in development; however, the level of processing at which this effect arises remains unclear. Here we show that the mapping occurs prior to conscious awareness of the visual stimuli. Under continuous flash suppression, congruent stimuli (e.g., “kiki” inside an angular shape) broke through to conscious awareness faster than incongruent stimuli. This was true even when we trained people to pair unfamiliar letters with auditory word forms, a result showing that the effect was driven by the phonology, not the visual features, of the letters. Furthermore, visibility thresholds of the shapes decreased when they were preceded by a congruent auditory word form in a masking paradigm. Taken together, our results suggest that sound-shape mapping can occur automatically prior to conscious awareness of visual shapes, and that sensory congruence facilitates conscious awareness of a stimulus being present.

Welcome to this Special Issue of Array: Proceedings of Si15, the 2nd International Symposium on S... more Welcome to this Special Issue of Array: Proceedings of Si15, the 2nd International Symposium on Sound and Interactivity.
The articles in the present issue originated in the Si15 Soundislands Festival, which was held in Singapore 18–23 August 2015. The festival events included five invited artist performances, two scientific keynotes and two days of proceedings, a commissioned sound installation, an afternoon of public talks, an internet panel, two pedagogic workshops, a concert with young performers, and more than fifty artworks and scientific papers in numerous forms and formats selected from an open call (http://soundislands.com/si15).
We are thrilled to present 20 articles, by 31 authors, emanating from Si15. The articles have been extended and thoroughly revised for this special issue of Array. They cover a range of topics related to aesthetics, percep-tion, technology, and sound art. We hope that you will enjoy the fruits of the authors' labour and therein discover many a stimulating thought.