Schizophrenia (SCZ) is a chronic, disabling, and debilitating disorder of the central nervous system that includes both psychotic and cognitive symptoms. It affects 1.0% of the population worldwide and is a public health concern (Mathers and Loncar, 2006; Jablensky, 2010). Patients with SCZ may have comorbidities, such as depression, anxiety, and substance abuse (Buckley et al., 2009; Sagud et al., 2009).

Studies that investigated substance abuse in patients with psychiatric disorders suggest that tobacco abuse may act as a form of self-medication for disease distress (Manzella et al., 2015). Chronic smoking is oftentimes comorbid with SCZ, which can affect visual sensitivity. Studies reported that patients with SCZ have a high rate of tobacco use and can be classified as heavy tobacco users (i.e., ≥25 cigarettes per day compared with <20% of the general population who smokes; Kelly and McCreadie, 2000). Patients who are diagnosed with SCZ may have a higher intensity of tobacco use and a lower rate of cessation compared with the general population (Beck et al., 2015).

Tobacco cigarettes contain numerous compounds that are harmful to health (World Health Organization [WHO], 2015). The most common form of cigarette absorption is through the main alkaloid in the cigarette, nicotine. Nicotine is a psychoactive ingredient in cigarettes that quickly reaches the brain through the airways (Balfour and Munafò, 2015). Nicotine binds to nicotinic acetylcholine receptors (nAChRs) in the central and peripheral nervous systems and upregulates the number of nAChRs (Levin, 2001; Metherate, 2004; Govind et al., 2009). The greater number of nAChRs is related to a greater need to use tobacco, thus strengthening addiction (Dani and De Biasi, 2001; D’Souza and Markou, 2011).

Schizophrenia is characterized by an imbalance of neurotransmitters, especially dopamine hyperfunction, and acetylcholine is involved in the ability of nAChRs to regulate dopaminergic activity (Levin, 2001; Blake et al., 2015). The activation of nAChRs increases dopamine release, and nicotine interacts with dopaminergic neurons that have cortical and subcortical projections (Levin, 2001; Balfour and Munafò, 2015).

Regardless of smoking status, SCZ patients present impairments in visual information processing, including motion perception, perceptual organization, spatial localization, temporal prediction, eye tracking, color perception, and visual contrast sensitivity (Yee et al., 1987; Shuwairi et al., 2002; Butler et al., 2008; Levy et al., 2010; Herzog and Brand, 2015; Martin et al., 2017). The results of studies that investigated changes in visual sensitivity in SCZ have been controversial. Generally, some authors argue that an imbalance in neurotransmission is responsible for a decrease in visual ability (Kéri et al., 2002; Chen et al., 2003; Cadenhead et al., 2013). Others argue that molecular dysfunction of visual processing pathways is both a cause and effect of visual impairment (Martínez et al., 2008; Gracitelli et al., 2013). As stated, losses in visual sensitivity in SCZ have been discussed in the literature (Silverstein and Rosen, 2015), indicating the existence of possible dysfunction of the magno- and parvo pathways. The M-pathway is highly sensitive to low spatial frequencies, whereas the P-pathway is highly sensitive to medium and high spatial frequencies. For example, Martínez et al. (2008) used functional magnetic resonance imaging and found that patients presented a decrease in M-pathway activity and a consequent decrease in the detection of low spatial frequencies (observed in the parietal and temporal lobe). When dysfunction of these pathways occurs, there may be losses in signal amplification, as reported by Kéri et al. (2000), Butler et al. (2005), Chen (2011), Schallmo et al. (2015), and others. This dual channel is further addressed in the section “Discussion.” However, other results suggest that overall dysfunction may be a more plausible approach than specific dysfunction of, for example, only the M-pathway (Herzog et al., 2013; Herzog and Brand, 2015). For a broad discussion between the magno- and parvocellular pathways, see Skottun (2015).

Likewise, heavy tobacco users without neuropsychiatric disorders, regardless of gender or age, also present spatiotemporal impairments (Kunchulia et al., 2014; Fernandes et al., 2017a). One may argue that this would occur in heavy tobacco users with SCZ. However, we need to consider that SCZ patients may use tobacco chronically for self-medication. The use of cigarettes in this population may reduce adverse effects or improve symptoms (Bakanidze et al., 2013). Investigating visual processing in patients with SCZ may serve as a noninvasive diagnostic tool to improve prognosis and quality of life (Silverstein et al., 2015).

The contrast sensitivity function (CSF) is a classic measure that was developed in the 1960s (Campbell and Maffei, 1974) to describe the mechanisms that are involved in spatial processing. The CSF is one of the main measures to verify the response of the visual system in healthy, preclinical, and clinical populations under scotopic, mesopic, and photopic luminance conditions. The CSF also allows observations of activity of the magnocellular, parvocellular, and koniocellular visual pathways (Xu et al., 2001; Ahmadi et al., 2015; Skottun, 2015). It also allows determination of the response of the visual system in processing rough attributes and fine details of objects that are modulated by low, medium, and high spatial frequency bands (DeValois and DeValois, 1990; Cornsweet, 2012).

Color vision-related losses can be either acquired or congenital (Lanthony and Dubois-Poulsen, 1973). Color vision loss can result from optical, neural, or systemic disease (Simunovic, 2016). Conditions that affect the nervous system, such as SCZ, may affect color vision (Nork, 2000; Costa et al., 2007; Fernandes et al., 2017b), ranging from ophthalmic diseases to pathology of the visual cortex. There is a large gap in the literature with regard to studies that evaluated color vision in SCZ. New tools and further studies are needed to completely assess this system. To avoid within-subject variability, the Cambridge Colour Test (CCT) allows the precise control of chromaticity parameters of stimuli, multiple randomized presentations of initial target-background chromaticity differences, and a staircase psychophysical procedure for estimating discrimination thresholds (Paramei, 2012). Thus, only one experimental session is sufficient to characterize color detection thresholds because repeated measurements are used. As one test of the CCT, the Trivector test is another way to assess congenital and acquired damage and can be performed quickly (i.e., <5 min). The advantage of a computer-controlled test is that the difference between the stimulus and background can be adjusted dynamically according to individual performance (Mollon and Regan, 2000). However, the Trivector test uses pseudoisochromatic stimuli that exhibit dynamic variations in chromaticity between the target and its background (Hasrod and Rubin, 2015; Fernandes et al., 2017c).

Considering these shortcomings, the main purpose of the present study was to investigate the effects of tobacco use on visual processing in patients who were diagnosed with SCZ. Understanding the relationship between visual processing and comorbid tobacco use and SCZ may have clinical applications. The present study evaluated the effects of tobacco use on visual sensitivity in SCZ patients. Our hypothesis was that chronic cigarette use affects the retina-geniculate-striate system through the inhalation of toxic smoke or the absorption of health-damaging particles that are found in cigarettes. We expected that tobacco users with SCZ would present greater impairments in visual sensitivity than SCZ patients who did not use tobacco.

The present study evaluated visual sensitivity in SCs compared with SCZ tobacco-user and -nonuser groups. The SCZ tobacco-user group had lower visual sensitivity than the SC group and SCZ tobacco-nonuser group for both chromatic and achromatic stimuli (Figures 1, 3). The CSF efficiently assessed possible effects of tobacco in SCZ patients, and these effects were modulated by low, medium, and high spatial frequency bands (Figures 1, 2). The CCT efficiently assessed discrimination thresholds for protan, deutan, and tritan axes (Figures 3, 4).

The CSF refers to the ability to detect variations in luminance. It is a relevant measure because of its low-level visual assessment, the impairment of which would suggest basic visual system disturbances in SCZ (Butler et al., 2008; Silverstein, 2016). Although it would be difficult to explain these disturbances in terms of cognition, a reduction of visual sensitivity could result from affective and cognitive deficits that are caused by the disease, such as illness severity, attention, working memory, planning, and psychomotor performance. We performed bootstrapping regression analysis to investigate this assumption. The results showed a relationship only between illness severity and visual performance, measured by the BPRS (Table 3). Investigating illness severity and illness duration may serve as a way to improve prognosis (Konarski et al., 2007). Non-response to treatment may lead to overall impairment, including a decrease in cerebral metabolism, which can cause atrophy in visual processing pathways (Wong-Riley, 2010).

With regard to color processing, trichromatic vision is based on the integration of short, medium, and long wavelengths by cone receptors. Another theory, the opponent process theory, implicates the existence of excitatory and inhibitory connections between cones (red–green and blue–yellow color vision (CV) systems; Goldstein, 2009). Both theories seek to explain optical and neural aspects of CV processing (DeValois and Webster, 2011). We observed a reduction of color discrimination thresholds in the SCZ group (Figure 3). More specifically, the SCZ tobacco-user group presented a significant reduction compared with the other two groups (Figures 4, 5), supporting the hypothesis that long-term smoking in schizophrenia may affect color vision.

Although the specificity of impairments that were observed in the magno- (M), parvo- (P), and koniocellular (K) pathways hinges on luminance conditions (Klistorner et al., 1997; Yoonessi and Yoonessi, 2011; Ahmadi et al., 2015), several authors concluded that spatiotemporal frequencies are associated with these pathways (Skottun and Skoyles, 2007; Cadenhead et al., 2013; Gracitelli et al., 2013; Skottun, 2015).

The dual channel model (involving the M- and P-pathway) is widely discussed in studies of visual psychophysics and schizophrenia. However, some studies reported an overall reduction of visual contrast sensitivity while isolating stimuli that would theoretically involve only one of these pathways (Skottun, 2000, 2015; Cadenhead et al., 2013). One may argue that this model is outdated and does not provide important descriptions of human visual function. Nonetheless, this model has provided insights into the function of chromatic and achromatic stimulus detection. For example, the M-pathway responds to achromatic stimuli with low spatial frequencies. The P-pathway responds to achromatic stimuli with high spatial frequencies and chromatic stimuli. Regardless of the luminance condition, both pathways can respond transiently, suggesting the importance of understanding these pathways separately to better understand overall function (Ahmadi et al., 2015; Skottun, 2015).

Under conditions that affect the central nervous system, such as SCZ and smoking, both the M- and P-pathways may exhibit dysfunction because of an imbalance of neurotransmitters (Kéri et al., 2002; Chen et al., 2003; Govind et al., 2009). For example, dopaminergic hypofunction that is caused by an imbalance of cortical upregulation may increase functional antagonism between the center and periphery of the receptive field of bipolar cells (Chaffiol et al., 2017; Nakao et al., 2018). The reverse may also be true, in which dopamine hyperfunction may decrease receptive fields of bipolar cells. Under light-adapted conditions, dopamine receptors are able to either decrease or increase the size of the receptive field (Zhang et al., 2011, 2012). Dopamine hyper- or hypofunction may involve changes that are related to spatial vision (Chen et al., 2003). Thus, abnormalities in these pathways may cause direct losses of the perception of stimuli at specific spatial or temporal frequencies. Nonetheless, there is a suggestion that an overall deficit of visual pathways may be more likely (Herzog and Brand, 2015).

Transient nicotine administration has been shown to improve cognition and visual backward masking (Levin et al., 1998; Heishman et al., 2010; Shaqiri et al., 2015), whereas long-term smoking may affect visual processing (Kunchulia et al., 2014; Fernandes et al., 2017a). The chronic use of cigarettes increases the expression of nAChRs in the nervous system (Besson et al., 2007), resulting in an imbalance of neurotransmission. Previous studies reported that chronic heavy smoking affected spatiotemporal vision in smokers (Kunchulia et al., 2014; Fernandes et al., 2017a). Therefore, in the present study, we opted to not include a group of healthy controls because this would not provide novel insights into this assumption.

As widely reported in the literature, there is a relationship between genetic variations of the nAChR α7 subunit (CHRNA7) and SCZ (Bakanidze et al., 2013). The deletion of 15q13.3 (i.e., the location of CHRN7 on the chromosome) may directly impact post-synaptic nAChRs (Deutsch et al., 2016). nAChRs are composed of α2–α10 and β2–β4 subunits, and one genetic variation of CHRNA7 may affect the response to chronic smoking (Dani, 2015). Furthermore, nAChRs are involved in modulating neurotransmission. Thus, the nicotinic system plays an important role in chronic smoking that is observed in SCZ patients. Genome-wide association studies of SCZ reported higher rates of smoking but failed to explain how this occurs (Vink et al., 2009; Gage et al., 2017). The function of the CHRNA5–A3–B4 cluster suggests a relationship between higher rates of smoking in SCZ patients (Lassi et al., 2016). A recent study reported evidence of a non-causal effect of tobacco use in SCZ (Lassi et al., 2016). The lack of a causal relationship between higher rates of smoking and schizophrenia may be related to the minor allele of the rs1051730 single-nucleotide polymorphism of the CHRNA5–CHRNA3–CHRNB4 region (Wium-Andersen et al., 2015). The aforementioned studies provided the impetus for conducting the present study.

The use of medications in SCZ seeks to improve disease symptoms. Pharmacological treatments usually involve antipsychotic medications, the function of which is to reduce psychotic states, decrease hallucinations, and improve symptomatology through the blockade of dopaminergic neurotransmission in the mesocortical and mesolimbic pathways (Lally and MacCabe, 2015). Typical antipsychotic drug treatment can impair visual processing. Several studies reported that atypical antipsychotics improved visual processing compared with typical antipsychotics (Chen et al., 2003; Cadenhead et al., 2013; Shoshina and Shelepin, 2015).

To homogenize the groups in the present study, only patients who took atypical antipsychotics were considered (Table 1). The CSF has been shown to decrease as the affinity of a drug class for dopaminergic neurotransmission increases (Chen et al., 2003; Cadenhead et al., 2013). In the case of atypical antipsychotics, dopaminergic affinity is lower. We employed visual psychophysics to determine whether the effects of smoking would be more evident than the effects of antipsychotic drugs. We found that smoking further affected contrast sensitivity beyond the effects of antipsychotics.

The SCZ tobacco-nonuser group was less severely impaired than the SCZ tobacco-user group, which may explain why these SCZ patients did not smoke. The SCZ tobacco-nonuser group also had better visual sensitivity. One limitation of the present study, however, was that the SCZ patients were not grouped according to illness duration. Chronic tobacco use in SCZ increases the activity of enzymes (e.g., CYP 1A2) that decrease the concentrations of antipsychotics (Sagud et al., 2009). According to Chen et al. (2003), the lack of use of medication would decrease contrast detection thresholds, thus increasing contrast sensitivity. The chronic use of cigarettes may decrease the concentrations of medications in the body. Therefore, the deleterious effects of antipsychotic on CSF may be significantly influenced by cigarette use.

One may argue that smoking affects vision, causing pupillary constriction (Lie and Domino, 1999), and our findings may be explained in view of this. However, the tests we conducted in the present study were performed at a relatively close distance. That is, for stimuli that were near the observer, pupillary constriction would still allow sufficient light energy distribution to detect the stimulus. Under photopic conditions, such as in the present study, and using tasks in which the observer is close to the computer monitor, pupillary constriction occurs automatically. Thus, even if cigarette smoking causes pupillary constriction, the relative energy that is necessary for detecting the stimulus is sufficient under photopic conditions (Alfonso et al., 2007). Schizophrenia may also be correlated with a decrease in pupillary diameter, which can affect the detection of complex stimuli (Minassian et al., 2004). However, the stimuli that were used in the present study were elementary in terms of visual processing and required minimal cognitive activity. Moreover, our regression analysis did not reveal a significant effect of cognition on visual performance. Nonetheless, we did not evaluate pupillary diameters pre- vs. post-smoking, which may be another limitation of the present study.

Our study replicates previous findings about impairments in visual processing in SCZ (Skottun and Skoyles, 2007; Cadenhead et al., 2013; Plomp et al., 2013; Silverstein and Rosen, 2015) and provides new directions for research on SCZ, tobacco use, and spatial vision. One promising approach could be to administer nicotine directly to SCZ tobacco-nonusers (e.g., using nicotine patches) to assess the CSF and other visual functions. Our study also highlights the need to evaluate nicotine-addictive behavior in patients with SCZ. Overall, our findings demonstrate the consequences of nicotine on the visual system in patients who are diagnosed with SCZ.

Despite the novelty of the present study, it also has several limitations. First, the findings would have been bolstered by specific measures of nicotine intake (e.g., serum cotinine and breath carbon monoxide). However, we attempted to control some variables in this regard, so we believe that this limitation did not affect our overall results. Second, the sample size may not have been sufficient to obtain reliable results in the regression analysis. Although a standard in psychophysical and perceptual studies, our sample size may be a limitation. Third, we cannot ascribe the differences that we observed to any specific visual pathway. Fourth, we did not randomize or evaluate the SCZ patients based on illness duration. Fifth, we used Snellen’s eye chart solely as a screening tool and not as an additional way of investigating visual acuity, which may also be a limitation. Our future studies will use logMAR rather than the eye cart as a screening tool. Sixth, the use of the BPRS instead of the Positive and Negative Symptom Scale (PANSS) may also be a limitation. However, we sought to verify whether the severity of SCZ symptoms, based on the BPRS, is a predictor of visual sensitivity. We will employ the PANSS in future studies.

The present findings need to be tested in clinical trials or larger controlled studies. The present study did not reveal a definitive origin of the losses in visual sensitivity. We employed a systematic methodology to evaluate visual sensitivity thresholds through the application of standardized tests with adequate levels of sensitivity and specificity. Such tests have been employed in previous studies of spatial vision.

In summary, the present results indicated significant differences in visual sensitivity in SCZ patients who were chronic tobacco users. Our findings justify future research on the pathophysiological mechanisms that are involved in these sensorial alterations.

TF: conceived and designed the experiments, participated in its coordination, performed the statistical analysis, and helped draft the manuscript. MA: helped revise the manuscript and interpret the data. JS: helped interpret the data and perform the statistical analysis. RN: conceived the experiments, helped interpret the data, and provided the statistical tools. NS: conceived and designed the experiments, helped draft the manuscript, and helped interpret the data. All of the authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.