Adolescents using combustible tobacco are 9 to 15 times more likely to use cannabis than non-smoking adolescents, while over half of U.S. adolescents between the ages of 12 and 17 years who smoke cigarettes report past-month use of cannabis . Co-use of cannabis and tobacco may interact, both acutely during co-administration and chronically over time, leading to complex immediate-, shorter-, and longer-term effects on cognition, brain, and behaviors. The co-occurrence of cannabis and tobacco use is concerning given its association with greater frequency of use and addiction severity, and poorer treatment outcomes related to both cannabis use disorders and tobacco use disorders. Although co-use of tobacco and cannabis is common among youth, little is known about the combined effects of combustible tobacco and cannabis on brain function and structure. Two groups independently found gray-matter volume differences in the putamen, thalamus, hippocampus, precentral gyrus, cerebellum, and prefrontal cortical regions between tobacco-using, cannabis-using and tobacco and cannabis co-using adults. Distinct and overlapping relationships with tobacco and cannabis measures and brain function and network connectivity at rest and during reward anticipation have also been described in TC subjects . Across studies, differences in brain volume and activation patterns between TC, mono-drug-using, and non-smoking subjects are most consistently observed in core regions and networks involved in cognitive control, attention, and reward processing. How brain activation patterns in these regions during reward processing relate to tobacco and cannabis grow supplies addiction severities is poorly understood and could reflect transdiagnostic or substance-specific processes in TC adolescents.
Understanding the potential effects of tobacco and cannabis on reward processing in TC adolescents has significant public health implications. Event-related potentials are well suited to evaluate mechanisms underlying reward processing during rapid decision-making . The feedback-related negativity , also termed reward positivity, feedback error-related negativity, and medial frontal negativity [see and for reviews], is an ERP component over mediofrontal areas of the scalp occurring between 200 and 300 ms after reward-related feedback and is observed during human trial-and-error learning and guessing tasks . Localized to the anterior cingulate cortex , it is described as the difference in ERP amplitude between positive and negative feedback and incorporates elements of valence, saliency, and expectancy . The FRN is sensitive to a reward prediction error signal that is generated when transient shifts in midbrain dopamine levels, in response to positive versus negative feedback of varying probabilities, signal disinhibitory neurons in the dorsal ACC . The FRN may emerge primarily from loss feedback and reflect a binary evaluation of good versus bad outcomes, with no difference between neutral and loss outcomes . This interpretation is based upon two lines of evidence. First, early studies of the FRN found it to be insensitive to the magnitude of reward and loss feedback . This insensitivity to magnitude has been called into question by multiple studies and a recent meta-analysis . Second, Holroyd, Hajcak, and colleagues observed that in EEG studies using trial-and-error learning or gambling tasks that included win, loss, and neutral conditions, no difference in amplitude was found between neutral and loss conditions . The binary function theory of FRN has not been tested in pediatric samples or examined developmentally. To date, few published studies have examined FRN in relation to SUDs, and results across studies have been mixed . Joyner and colleagues recently examined FRN in relation to SUD problems in a large sample of adults and found that FRN, measured as the net difference between win- and loss-related activation, was negatively correlated with SUD symptomatology . Our group has published two studies examining the FRN in high-risk adolescents both aligning with the reward deficiency model of addiction vulnerability.
Adolescents who had been prenatally exposed to cocaine demonstrated decreased FRN amplitude in response to losses compared to gains when compared to matched controls . Yau and colleagues observed a blunted feedback for both win and loss conditions during a risk-taking task in adolescents with at-risk or problematic internet use . No studies to date have examined the FRN in tobacco-using or TC adolescents. Here we examined differences in mediofrontal electrocortical activity elicited by monetary reward, neutral, and loss feedback conditions, indexed by the FRN, in relation to cannabis-related and tobacco-related problem severity in adolescents with biochemically verified daily tobacco smoking who regularly use cannabis and tobacco, and a matched group of non-smoking healthy control participants. We predicted that the FRN amplitude would differentiate between reward and non-reward outcome, with no difference between neutral and loss feedback, consistent with FRN studies in adults . Based upon previous feedback-related ERP studies in high-risk youth and substance-using adults, we hypothesized that FRN amplitude across feedback conditions would be decreased in tobacco-smoking adolescents compared to controls. We also predicted that cannabis- and tobaccorelated problem severity would be negatively correlated with FRN amplitude among smoking adolescents. Earlier ERP studies of feedback processing in samples of high-risk adolescents and adults with SUDs have not reported latency outcomes; thus, we had no direct data to inform our latency hypotheses. Based upon indirect evidence of opposing effects on orientation and processing speed from acute cannabis and tobacco administration , we anticipated seeing shorter FRN latencies in relation to tobacco use and longer FRN latencies in relation to cannabis use. A telephone interview was administered to adolescents and their parents/guardians prior to study entry. Participants who met inclusionary criteria, and whose parents provided consent if under age 18 years, were then scheduled for a single 3-hour study session. In the session, participants completed self-report questionnaires, behavioral assessments, biochemical measures, and the EEG scan.
For smoking adolescents, inclusion criteria included current daily cigarette use and current or past history of smoking 5 or more cigarettes on a daily basis for at least a 6-month period, urine cotinine level above 500 ng/ml at study visit, no current illicit substance use and a urine drug screen negative for drugs other than cannabis. For HCs, criteria included never smoking daily, no history of regular patterns of smoking, urine cotinine level lower than 100 ng/ml at study visit, no history of illicit substance use , and not meeting criteria for heavy drinking . For all participants, criteria included ages 14–21 years, English language fluency, full scale IQ > 70, no chronic medical illnesses, no evidence of serious mental illness , no history of lifetime or current DSM-IV-TR diagnosis of dependence on another psychoactive substance . Additional exclusion criteria included neurological conditions , head trauma with loss of consciousness > 2 min, use of any psychoactive drugs including anxiolytics and antidepressants unless the adolescent had been taking the medication consistently for 3 months, and pregnancy or lactation. Participants provided consent/assent, and participants under age 18 years also had a parent/guardian provide consent. This study was approved by the Yale University School of Medicine Human Investigation Committee. All participants were instructed to abstain from alcohol or drugs other than cannabis or tobacco for 24 h on scan days. Participants were not instructed to modify their cannabis grow facility and tobacco use but were informed that if they presented for the scan day showing signs of overt intoxication that they would be rescheduled. Smoking participants were given an opportunity to smoke a tobacco cigarette prior to initiating study procedures. All participants were asked their last day and time of use cannabis, tobacco, and alcohol, assessed for signs/symptoms of intoxication, and were tested for recent drug and alcohol use and for expired carbon monoxide levels via breathalyzers and urine biospecimen collection. From the urine biospecimen, three biochemical measures were obtained: the presence of drugs of abuse were assessed via a qualitative UDS; urine cotinine levels were assessed via a semiquantitative urine cotinine test ; and quantitative urine cannabinoid level were assessed via mass spectroscopy . Although there were four options on a given trial, feedback was rigged to have the probability of 33.3% reward, 33.3% neutral, and 33.3% loss across the task. Feedback was random, meaning that there was no pattern of certain balloons predicting specific outcomes, but adolescents were led to believe that some people ‘can figure out a pattern some of the time’. Participants were reminded to look at the screen and not at their hands, as they would in a video game to reduce eye-movement artifact.
Participant earnings were displayed numerically on the screen, centered just below the middle two balloons. There were four blocks of trials with approximately 45 trials in each block. After each block, a clear glass coin jar appeared to reflect cumulative winnings to that point. Realistic quarter images appeared in the jar, one by one, each followed by a coin sound. Prior to beginning the game there were 3 practice trials, which introduced the game and coin jar. A total of 180 trials were administered for the purpose of computing ERPs. Total winnings from the ERP reward-feedback game were $7.25 for each participant. Participants received this payment as part of a larger fixed compensation for completion of the whole study. Each participant was seated 24 in. in front of a 19 in. computer LCD monitor. Each participant’s head circumference was measured to determine the appropriate net size and to mark the Cz as the juncture of the halfway point between naison to inion and left and right preauricular notches. Next, a Hydrocel high-density array of 128 Ag/AgCl electrodes arranged into a net was placed on the participant’s head using standard procedures. Before this, the net was soaked in warm potassium chloride solution that served as the electrolyte.Brain wave data were recorded using the Netstation v.4.4 software package and EGI high impedance amplifiers, sampling at 1000 Hz . The online filters were set at 0.1–1000 Hz. All electrodes were referenced to Cz for recording and then re-referenced offline for data analysis. All impedances remained at or under 40 kΩ as indicated by impedance measures made immediately before and after the test session. The E-prime v.2.0 software package controlled the stimulus presentation. Each participant’s EEG and behavior were continuously monitored across the session so that stimulus presentation occurred only when the participant was sitting still and looking at the monitor. Offline post-processing occurred in the Netstation v.4.4 software package The EEG data were first processed through a 0.3 Hz first-order high-pass filter and a 30 Hz low-pass filter. Then they were segmented to epochs that contained a 100 ms pre-stimulus baseline and 600 ms post-stimulus interval. Bad eye channels were manually marked and interpolated by surrounding channels. In the next step, artifact rejection was applied, in which bad segments were marked. Epochs with any eye blink or eye movement were rejected. Epochs with more than 10 bad channels were rejected as well. Then the remaining bad segments were replaced by surrounding channels. The single trial data were re-referenced from the vertex to an average reference of all electrodes because the latter was thought to be a better representation of true zero .
The data were baseline-corrected to a 100 ms pre-stimulus interval. Finally, single-trial data were averaged respectively for each condition . Participants providing at least 30 artifact-free trials per condition were included . Data for participants with fewer than 30 artifact-free trials per condition received additional preprocessing with statistical eye-blink removal.Participants whose data yielded 30 good trials per condition with this additional approach were then included in the overall statistical analysis . The participants receiving artifact removal were not significantly different from those not receiving artifact removal in terms of reward/neutral/loss ERP amplitude, latency, age, sex, IQ, or group status. Past work on the feedback negativity has localized the FRN to the medial frontal region along the midline at site Fz . We relied on the average signal of four electrodes over the midline in this region, specifically electrode numbers 11 , 12, 5, and 6 consistent with prior studies . For ERP analysis, the FRN amplitude was defined as the mean ± 25 ms around the negative peak amplitude between 200 and 350 ms within our electrode cluster. Latency for the negative peak of the FRN was assessed over the same channels and in the same 200–350 ms window.