Researchers use fMRI and state-of-the-art brain mapping techniques to study alcohol's effects on first-year students
Brain regions of interest (green) and group-level brain activity (red-yellow); data were averaged across three fMRI sessions and overlaid on a standard brain template. Credit: Penn State Social, Life, and Engineering Sciences Imaging Center (SLEIC)
By: Seth Palmer
This article is the first of a series exploring the diversity of interests and variety of experimental approaches represented by neuroscientists at the Huck Institutes.
Investigating subjects ranging from neural response and habituation to alcohol, to mechanisms of motor control, to effects of neural injury, these researchers are devising unique methodologies employing a wide variety of technologies and techniques — applying statistical network analysis to data from functional MRI scans, using electromyography in tandem with neuronal tracers and fluorescence microscopy, and combining MRI and brain lesion analysis with virtual reality experiments — and they are making discoveries with the potential to change the way we experience our world: new understanding of how alcohol cues influence the brain, which could make college students think differently about social drinking; newfound parallels between our own neural processes and those of other organisms, which could help to unlock neural code to drive brain-computer interfaces; and new insights into the effects of specific neural injuries on sensorimotor processes, which could lead to developing more-effective post-stroke rehabilitation.
If you find this all to be a little mind-boggling, you're not the only one! Groundbreaking neuroscience such as this makes it clear that the possibilities are, fittingly, as limitless as the imagination.
Anecdotal evidence abounds attesting to the many negative social and physical effects of what has been called the "freshman binge" — the dramatic increase in alcohol use that comes with many students' first year of college — and the behavioral changes that accompany them.
Such changes in behavior are indicative of underlying changes in the brain, but in contrast to alcohol's numerous other effects, its effect on the brain's continuing development from adolescence into early adulthood — which includes the transition from high school to college — has been significantly less studied.
So with a team of researchers, graduate student Adriene Beltz set out to investigate the changes that occurred to alcohol-related neural processes in the brains of a small group of first-year students.
Using functional magnetic resonance imaging (fMRI) and a data analysis technique known as effective connectivity mapping — done with an advanced approach known as uSEM and a state-of-the-art statistical program, recently developed at Penn State, called GIMME — the researchers collected and analyzed data from eleven students, who participated in a series of three fMRI sessions beginning just before the start of classes and concluding part-way through the second semester.
"We wanted to know if and how brain responses to alcohol cues — pictures of alcoholic beverages in this case — changed across the first year of college," says Beltz, "and how these potential changes related to alcohol use. Moreover, we wanted our analysis approach to take advantage of the richness of fMRI data. By employing effective connectivity mapping at the individual level, our findings reflect the spatial and temporal dynamics of key brain networks, and the similarities and differences of those networks across students."
Results of the study
Analysis of the data collected from the eleven study participants revealed signs in their brains' emotion processing networks of habituation to alcohol-related stimuli, and noticeable alterations in their cognitive control networks.
While these findings may seem somewhat intuitive (i.e., "Yes, I'm sure that alcohol changes my brain in some not-so-nice ways..."), their implications for students' continuing development may not be.
Many of us might be inclined to think that young adults' cognitive development is, for the most part, complete by the time they enter college, but recent studies have indicated that the brain's development continues through our mid-twenties, particularly in those regions of the brain responsible for decision-making or judgment-related activity — the sort of cognitive "fine tuning" that potentially makes us, in some senses, as much who we are (and will be) as any other stage of our overall development.
Other recent studies suggest that binge drinking during late adolescence may damage the brain in ways that could last into adulthood.
Beltz's study indicates that connections among brain regions involved in emotion processing and cognitive control change with increased exposure to alcohol and alcohol-related cues, but could this also influence other parts of the brain, such as those still-developing regions responsible for students' decision-making and judgment abilities? And could those effects potentially last a lifetime?
"The brain is a complex network," Beltz explains. "We know that connections among different brain regions are important for behavior, and we know that many of these connections are still developing into early adulthood. Thus, alcohol could have far-reaching consequences on a maturing brain, directly influencing some brain regions and indirectly influencing others by disrupting neural connectivity."
A unique methodology
The study is a first-of-its-kind pilot, based on a unique methodology.
It is a longitudinal study — meaning that the data were collected based on the same set of observations made over an extended period of time — and the researchers used functional magnetic resonance imaging (fMRI) along with several state-of-the-art analysis tools to create what are known as effective connectivity maps, illustrating patterns of activity in the participants' brains.
The fMRI scanner measures a marker of neural activity called the blood oxygen level-dependent (BOLD) contrast — differences (contrast) in magnetization due to changes in blood oxygen levels that occur as a result of neuronal activity; the fMRI data can then be processed using specialized software to create images of this activity in the brain.
Researchers like Adriene Beltz are particularly interested in the connectivity among different regions of the brain — how they communicate during a given task — and whether there are meaningful differences in this connectivity across individuals or groups of people.
"In this study," says Beltz, "we wanted to do more than describe which brain regions were involved in responding to alcohol cues. We wanted to show how those brain regions 'talk' to each other, both concurrently and at future time points, and how patterns of 'talking' differ across people."
By applying statistical tools such as uSEM (unified structural equation modeling) to longitudinal sequences of fMRI data, researchers are able to analyze changes in neural activity and map the connections that are made among different brain regions of interest (ROIs) — creating what are known as connectivity maps; when information is incorporated about the time course of connections among ROIs, the result is known as an effective connectivity map.
A problem arises, however, when individuals' data are aggregated to create group-level connectivity maps; connections sometimes appear for the group that did not exist for any individual. In order to address this problem, Kathleen Gates — a co-author of the study — created a program called GIMME (Group Iterative Multiple Model Estimation), which is able to incorporate individuals' unique variations into group connectivity maps, and thus avoid creating false connections for the group.
"In order to assess changes in alcohol-related brain processes across the first year of college," Beltz explains, "we needed to combine longitudinal data collection with data analysis techniques that are sensitive to individual-level brain processes; implementing uSEM through GIMME allowed us to do this. Other mapping techniques for fMRI data only consider concurrent brain connections at the group level, and they lose information about time and individuals."
Collecting and analyzing the data
Eleven first-year students — six women and five men, all between the ages of eighteen and nineteen — participated in the study, and provided data in three sessions: before the start of classes, during the first semester, and part-way through the second semester.
While in an fMRI scanner at the Penn State Social, Life, and Engineering Sciences Imaging Center (SLEIC), the students completed a task: responding as quickly as possible — by pressing a button on a grip device — to an image of either an alcoholic beverage or a non-alcoholic beverage when both types of images were displayed consecutively on a screen. From the resulting data, effective connectivity maps were created for each individual and for the group.
Examining the final maps, the researchers found that brain regions involved in emotion processing showed less connectivity when the students responded to alcohol cues than when they responded to non-alcohol cues, and that brain regions involved in cognitive control showed the most connectivity during the first semester of college — suggesting that the students needed to heavily recruit brain regions involved in cognitive control in order to overcome the alcohol-associated stimuli when instructed to respond to the non-alcohol cues.
"Connectivity among brain regions implicated in cognitive control spiked from the summer before college to the first semester of college," says Beltz. "This was particularly interesting because the spike coincided with increases in the participants' alcohol use and increases in their exposure to alcohol cues in the college environment. From the first semester to the second semester, levels of alcohol use and cue exposure remained steady, but connectivity among cognitive control brain regions decreased. From this, we concluded that changes in alcohol use and cue exposure — not absolute levels — were reflected by the underlying neural processes."
Implications and future directions
Although the immediate implications of the pilot study for first-year students are fairly clear (i.e., "You'll be doing yourself a favor if you lay off the booze!"), there are still a number of unanswered questions related to alcohol's longer-term effects on development — both for college students after their first year, and for those same individuals later in life.
To begin exploring those potential long-term effects, Beltz has planned a follow-up study to track a larger number of participants over a greater length of time.
"We are currently investigating whether students' alcohol use history early in college influences their brain function late in college," says Beltz. "Together, these two studies will highlight the importance of the college transition for short-term alcohol-related brain processes, and for long-term cognitive function."
“Adriene is extremely productive and capable in her research,” says her mentor, Sheri Berenbaum. “This study illustrates how impressive her work is: it is interdisciplinary and collaborative, employing sophisticated and innovative methods and demonstrating her ability to think critically and creatively about important research questions. I anticipate the same high quality in her dissertation research and in her subsequent career.”
Funding for this research was provided by NIAAA grant R01AA015737 to Rob Turrisi, instrumentation funded by the National Science Foundation through grant OCI-0821527, Penn State Institute of the Neurosciences, Penn State Social Science Research Institute, and Penn State Social, Life, and Engineering Sciences Imaging Center 3-Tesla Magnetic Resonance Imaging Facility.
More about the research team
- Kathleen Gates was Research Associate Professor of Human Development and Family Studies at Penn State at the time of the pilot study; she is currently Research Assistant Professor at Arlington Innovation Center, Virginia Polytechnic Institute and State University.
- Carmen Pulido is Assistant Adjunct Professor of Psychiatry at the University of California, San Diego.
- Robert Turrisi is Professor of Biobehavioral Health at Penn State.
Huck Institutes affiliates
- Adriene Beltz is a graduate student in the Psychology program at Penn State, and is advised by Sheri Berenbaum.
- Anna Engels is Assistant Director of the Social, Life, and Engineering Sciences Imaging Center (SLEIC) at University Park.
- Peter Molenaar is Distinguished Professor of Human Development at Penn State, and is a researcher in the Center for Brain, Behavior and Cognition as well as the Penn State Institute for Diabetes and Obesity.
- Sheri Berenbaum is Professor of Psychology and Pediatrics at Penn State, and is a co-funded faculty member of the Huck Institutes' graduate program in neuroscience.
- Rick Gilmore is Associate Professor of Psychology at Penn State, Director of Human Imaging at SLEIC, a researcher in the Center for Brain, Behavior and Cognition, and a faculty member of the Huck Institutes' graduate program in neuroscience.
- Stephen Wilson is Assistant Professor of Psychology at Penn State, and is a faculty member of the Huck Institutes' graduate program in neuroscience.
- Gates KM
- Pulido C
- Turrisi R
- Changes in alcohol-related brain networks across the first year of college: A prospective pilot study using fMRI effective connectivity mapping
- Addictive Behaviors 38(4): 2052-2059