Research paper for PSY 506 Fall 2004 (Return to classweb page)

 A summary of: "Arousal Modulation of NeonatalVisual Attention: Implications for Development"

by Leslie Swiney

When infants are first born, their vision is based on a balancebetween attention and arousal due to the lack of visual experience oftheir world around them. As people grow older, their perceptionexpands due to their experiences, which in turn allows them tointerpret visual information differently than when they were younger. Many different researchers have been interested in how visualattention differs from normal infants (normal referring to low riskinfants) to infants labeled as high risk (meaning substance addictedinfants or infants with Central Nervous System damage). The reviewedstudy by Gardner, Karmel, and Flory (2003) looked longitudinally atinfants' visual attention from the neonatal stage to two or moreyears of age.

Before looking at Gardner et al. (2003) study, many other studieshave been done over similar topics to visual attention in infancy. Some of the studies look at visual attention using physiologicalmethods (e.g. Richard, 2000 & Diamond, 2000). The physiologicalmethods allow researchers to see neural and cortical changes in theinfant while attending to a visual stimulus. Also, Singer (1999)claims that the only way to truly understand visual attention inhigh-risk infants is to improve the design of the experiment. Gardner et al., in response to Singer, developed a design that looksat internal and external stimuli together to create a more valid andreliable study. Other researchers (Jankowski, Rose and Feldman,2001; Frick and Richards, 2001) look behaviorally at visual attentionin infants to determine if there is any change over the course of thefirst few years of life.

Gardner et al. (2003), for the different stages of their research,were interested in arousal and attention in neonates that had CentralNervous System (CNS) damage. To find this population, theresearchers used infants who were assigned to the neonatal intensivecare unit and infants exposed to toxic drugs (such as cocaine). Gardner et al. compared the at-risk group (those with CNS damage) toa low-risk population (healthy infants) to see any differencesthroughout the longitudinal study. They first hypothesized thatdistress to the CNS would upset the system that controls forattention and arousal. They also hypothesized that the developmentof attention and arousal would differ depending of the type of CNSdamage.

Gardner et al. (2003) looked physiologically at the auditorybrainstem evoked responses (ABR) in neonates (0-9 months). Theyfound that neonates with CNS damage had irregular ABR waves thatparalleled adults who had suffered a head injury. After measuringthe ABR, the researchers then analyzed the behavioral effects ofsubstance-addicted (mainly cocaine-exposed) neonates. They foundsimilar results as the neonates with CNS damage; the infants ABRswere irregular. Richard (2000) used similar methods on low-riskinfants. He wanted to trace shifts in attention in young infantsusing scalp recorded event-related potentials (ERP). Richard hadinfants look at a computer screen where they were presented with a"cued target." The results showed that infants' looking increasedwith the amount of cortical development. This would explain theGardner et al. findings of why cocaine addicted and CNS damagedinfants have more trouble looking at novel objects; their corticalareas are still underdeveloped compared to normal infants.

Neonates that are born addicted to cocaine have many behavioral aswell as physiological problems. Singer (1999) claimed that allprevious research does not revealed any new information because ofthe small sample sizes. She believes that the small sample sizesmake it difficult to separate the large number of risk factors. While it is difficult to separate an infant's actual behavior fromthe behaviors that are caused by cocaine, much could be done if thedesign was changed. All research that has been done to date hasrelied on standard repeated measures. Gardner et al. (2003)understood the same method issues and this is the reason theyincorporated a larger sample population and studied specificbehaviors that may be linked to attention and arousal.

To study the neonates behaviorally, as well as incorporate a gooddesign, Gardner et al. (2003) gave each neonate a pair of stimuli atthree levels of arousal that correspond with feeding. The levelsconsisted of (1) being less aroused, after feeding, (2) more aroused,increase in internal stimulation (hunger), and (3) more aroused,increase in external stimulation (distracting environment). (Feedingis used because infants become more relaxed when they are full,therefore making them less aroused.) Gardner et al. found bothlow-risk and high-risk (both CNS injured and cocaine exposed) infantswere more likely to look at a novel stimulus after feeding (lessaroused) than before feeding. However, low-risk (normal) infantslooked longer at novel stimuli both before and after feeding ascompared to the high-risk infants. Meaning the low-risk infantsshowed greater focused attention than the high-risk infants. Thisstudy can conclude that high-risk infants have a more difficult timefocusing attention compared to low-risk infants. The importance ofadding the feeding factor is the researchers are able to narrow downthe external stimulation and relate low arousal to being hungry.

Once the infants began to get older (10-16 months), Gardner et al.(2003) analyzed the infants' higher-level processing of attention andbehavior. They were the most interested in if behavioral problems(such as focused attention) are due to being a high-risk infant orare they related to environmental issues. Gardner et al.hypothesized that infants who had CNS damage or cocaine exposure andwho had difficulty focusing on novel stimuli would have problemsregulating attention in a stimulating (distracting) environment. Theinfants were video taped for two-minutes at three different times. Each infant was presented with three different stimuli (toys) by aresearcher at the beginning of each trial. On a computer screen inthe same room as the infant, a distracter was presented for 4 secondsduring randomly assigned intervals of 3, 5, or 7. Each distracterwas presented with a beep. The researchers choose the intervaldepending on the way the infant was responding to the distracter.

When all of the infants were at 10-months, they all were morefocused on the toy rather than being upset by the distracter (bothfor low- and high-risk infants). However, the low-risk infants wereable to hold their focused attention longer on the toy than thehigh-risk infants. This result is similar to those at the neonateage when the low-risk infants were able to hold their focusedattention longer than the high-risk infants. The infants at 16months showed a more distinct difference in focused attention. Thelow-risk infants were even less distracted than they were at 10months, meaning they were able to focus their attention more stronglyon the toy rather than the distracter. The high-risk infants showedno change from 10 months to 16 months.

Another study done by Diamond (2000) looked at the relationshipbetween motor development and cognitive development in thecerebellum, which highly correlates to the Gardner et al. (2003)study. Diamond's study noted that motor coordination issues havebeen found in children with Attention Deficit Hyperactivity Disorder(ADHD). If motor abilities are affected by ADHD, then cocaineexposed and CNS damaged infants should also have difficultlycontrolling their motor skills while completing a task. Diamond alsonoted that in order for an infant to have proper cognitivefunctioning, the infant must concentrate. The best way to get aninfant to concentrate on a cognitive task is to make the task novel. Gardner et al. found their results to be similar to Diamond'sfindings; high-risk infants had a more difficult time staying stilland keeping their attention focused on the toy rather than thedistracter. Also when looking at attention and arousal, Gardner etal. found that infants concentrated more on novel stimuli.

After running their longitudinal research, Gardner et al. (2003)made a model of how predictable arousal and attention are based ontheir previous measures. According to their model, 0-1 month is ahigh predictor for 16 months infants' visual attention. However,there is a negative predictability from 0-1 month to 4 months and alow predictability from 0-1 month to 7-13 months. The researcherstried to clarify these differences through two explanations. First,there may have been "lagged effect," meaning there is an attentionand arousal delay between the ages of 0 months to 16 months. Anotherpossible explanation is their measurements for attention and arousalmay have differed slightly during the different ages of the infants. Thus, they may have measured other components of attention than whatGardner et al. were originally interested in between 0-1 months to 4months and 0-1 months to 7-13 months. This is one of thedifficulties when working with infants. Because infants cannotspeak, the researcher is drawing conclusions based fully on theirinterpretations of the physiological and behavioral components. Looking at visual attention in adults is easier because theresearchers can additionally assess his/her self-report of thestimulus.

Another piece of information Gardner et al. (2003) found due totheir earlier findings of this researcher was a relationship betweeninfant Arousal Modulated Attention (AMA) and language problems at 4-7years of age. Meaning, infants who have trouble focusing attentionhave shown language issues in later stages of development. Theyalso concluded that because behavior can be so complex, changes inbehavior might be due to external situations rather than just CNSdamage. From their results, Gardner et al. believe that behavior andattention (or attention and arousal) have less to do with just CNSdamage in the first year of life, but rather the degree to which theCNS damage is restored to health. The same is true with infants whoare exposed to cocaine while still in utero. Jankowski, Rose, andFeldman (2001) found that when infants look at a stimulus in a morebroad way (a wider area rather that just a single object); theinfants show better attention for that stimulus. The purpose of theJankowski et al. study was to see if attention changes over time. Jankowski et al. found that attention is malleable in infants andtheir perception of broader stimuli change over time. So Gardner etal. could be right to assume that attention to a stimulus couldchange with a lesser degree of CNS damage. The infant's attention isgoing to change as their experiences change.

Frick and Richards (2001) also believe that attention can change. They looked at infants' visual attention at 14, 20, and 26 weeks ofage in low-risk (normal) and high-risk infants (infants with heartdifficulties). Over this time period, Frick and Richards showeddifferent types of stimuli and varied the time the stimuli werepresented. They found that each infant was different; some infantslooked longer at a stimulus than others. They concluded thatinfants' visual attention is individually based and that there isreally no way to determine why some infants look longer than others. Frick and Richards study help validate Gardner et al. (2003)conclusions as well. Although Gardner et al. did find a differencebetween the low- and high-risk young infants, if they looked at theinfants in later life, they too may find that attention isindividually based. While more research needs to be done of thisarea, it is always going to be hard to separate behavior (externalstimuli) from physiological (internal stimuli) circumstances.


Diamond, A. (2000). Close interrelation of motor development andcognitive development and of the cerebellum and prefrontal cortex.Child Development, 71, 44-56.

Frick, J. & Richards, J. (2001). Individual differences ininfants' recognition of briefly presented visual stimuli. Infancy,2(3), 331-353.

Gardner, J., Karmel, B. & Flory, M. (2003). Arousal modulationof neonatal visual attention: Implications for development. In S.Soraci and K. Murata-Soraci (Eds.), Visual Information Processing(pp.125-153). Westport: Praeger.

Jankowski, J., Rose, S. & Feldman, J. (2001). Modifying thedistribution of attention in infants. Child Development, 72(2),339-352.

Richard, J. (2000). Localizing the development of covert attentionin infants with scalp event-related potentials. DevelopmentalPsychology, 36, 91-108.

Singer, L. (1999). Advances and redirections in understandingeffects of fetal drug exposure. Journal of Drug Issues, 29(2),253-262.