**2. Schizophrenia and cognitive biases**

As mentioned above, some cognitive biases have been widely related to schizophrenia such as need for closure [21], overconfidence bias [14], bias against confirmatory evidence (BACE), bias against disconfirmatory evidence (BADE) and above all jumping to conclusions [13, 17]. Then we will explain these four cognitive biases and their relationship with this disorder.

#### **2.1. Jumping to conclusions, against disconfirmatory evidence bias (BADE) and bias against confirmatory evidence (BACE)**

The jumping to conclusions is a cognitive bias where there is a data‐gathering bias and a contrast of hypothesis testing bias. This cognitive bias occurs when there is a tendency to make decisions very quickly even when there may not be a lot of evidence [22]. This bias is usually tested using probabilistic reasoning tasks based on a Bayesian model of probabilistic inference [13, 23]. As it is stated above, jumping to conclusions bias is a bias which has been largely related to patients with schizophrenia, especially, in patients with delusions [13, 17].

Numerous researches said that jumping to conclusions would be an endophenotype of psychosis [24]. In support of this, Menon et al. [25] found that antipsychotic treatment in patients with schizophrenia did not reduce the bias which may suggest that jumping to conclusions bias could be a trait maker for schizophrenia. This bias has also been observed in first‐degree relatives of patients with schizophrenia above all patients with delusions [26]. Similarly, these cognitive biases have been observed in patients with psychosis.

The jumping to conclusions bias is closely related to the origin and maintenance of delusions. In fact, most studies have found this bias in patients with schizophrenia who have positive symptoms and not in patients with schizophrenia who have a negative symptomatology. Therefore, this bias is more associated with the delusions rather than with schizophrenia disorder [26]. The JTC is found in people with delusions with a schizophrenia diagnosis or delusional disorder. The subjects with delusions show a bias in the collection of information. They need fewer data than the normal population to reach to a final decision. In delirium process, there is a development constant by which ideas confirm and disconfirm through approaches to hypothesis, information gathering and contrast of the results with the previous hypothesis. Probably some of these steps are inadequate in delusional patients, both cognitive and emotional processes.

At present, two different hypotheses have been advanced to explain this cognitive bias [27]. On the one hand, several authors support the hypothesis that people with jumping to conclu‐ sions bias overestimate the conviction in their choices at the beginning of the decision process [28]. In line with this viewpoint, patients with schizophrenia tend to accept choices early that wrong inferences may.

On the other hand, other authors argue that the bias could be due to a low information threshold for acceptance of a decision [29]. "The hypothesis of liberal acceptance" was proposed by Moritz and Woodward [30]. These authors based on their hypothesis in the decrement of confidence gap in patients with schizophrenia who made final decisions with little evidence collected for them. Moreover, liberal acceptance is thought as core deficit because it is exhibited in delusion‐relevant scenarios and neutral settings. Different experi‐ mental tasks have been used to support their hypothesis. For example, Moritz et al. [31] carried out an experiment called "Who wants to be a millionaire?," a TV game, in patients with schizophrenia and healthy control where they were asked to rate the probability of each of four response alternatives to general knowledge questions. The results showed that patients reached to final decisions at 54 % subjective probability ratings and healthy controls at 70 %.

different biases where the patient knows the bias, becomes aware of it and works on its

As mentioned above, some cognitive biases have been widely related to schizophrenia such as need for closure [21], overconfidence bias [14], bias against confirmatory evidence (BACE), bias against disconfirmatory evidence (BADE) and above all jumping to conclusions [13, 17]. Then we will explain these four cognitive biases and their relationship with this disorder.

**2.1. Jumping to conclusions, against disconfirmatory evidence bias (BADE) and bias against**

The jumping to conclusions is a cognitive bias where there is a data‐gathering bias and a contrast of hypothesis testing bias. This cognitive bias occurs when there is a tendency to make decisions very quickly even when there may not be a lot of evidence [22]. This bias is usually tested using probabilistic reasoning tasks based on a Bayesian model of probabilistic inference [13, 23]. As it is stated above, jumping to conclusions bias is a bias which has been largely

Numerous researches said that jumping to conclusions would be an endophenotype of psychosis [24]. In support of this, Menon et al. [25] found that antipsychotic treatment in patients with schizophrenia did not reduce the bias which may suggest that jumping to conclusions bias could be a trait maker for schizophrenia. This bias has also been observed in first‐degree relatives of patients with schizophrenia above all patients with delusions [26].

The jumping to conclusions bias is closely related to the origin and maintenance of delusions. In fact, most studies have found this bias in patients with schizophrenia who have positive symptoms and not in patients with schizophrenia who have a negative symptomatology. Therefore, this bias is more associated with the delusions rather than with schizophrenia disorder [26]. The JTC is found in people with delusions with a schizophrenia diagnosis or delusional disorder. The subjects with delusions show a bias in the collection of information. They need fewer data than the normal population to reach to a final decision. In delirium process, there is a development constant by which ideas confirm and disconfirm through approaches to hypothesis, information gathering and contrast of the results with the previous hypothesis. Probably some of these steps are inadequate in delusional patients, both cognitive

At present, two different hypotheses have been advanced to explain this cognitive bias [27]. On the one hand, several authors support the hypothesis that people with jumping to conclu‐ sions bias overestimate the conviction in their choices at the beginning of the decision process [28]. In line with this viewpoint, patients with schizophrenia tend to accept choices early that

related to patients with schizophrenia, especially, in patients with delusions [13, 17].

Similarly, these cognitive biases have been observed in patients with psychosis.

reduction and avoidance.

98 Schizophrenia Treatment - The New Facets

**confirmatory evidence (BACE)**

and emotional processes.

wrong inferences may.

**2. Schizophrenia and cognitive biases**

In addition, several studies have explored the possible relationship between jumping to conclusion bias and the bias against disconfirmatory evidence (BADE) and bias against confirmatory evidence (BACE) [13, 17].

Bias against disconfirmatory evidence (BADE) is a cognitive bias where, regardless of the inconsistent information, the hypothesis holds despite evidence to the contrary. Conversely, in the bias against confirmatory evidence (BACE) individuals, regardless of inconsistent information, they maintain their belief or hypothesis because of the evidence in favor of it [32]. The dimensional model of schizophrenia predicts unifying cognitive biases or combined cognitive biases to contribute toward the formation of the delusional aspects of psychosis [17]. For example, according to Munz [33], jumping to conclusions would play a facilitating role in the formation of new delusional systems and BADE. This dimensional model emphasizes quantitative gradations of psychopathology, both within and between subjects, rather than qualitative, discrete, all‐or‐none class distinctions. However, nowadays the relationship between these remains unclear because the studies are controversial and there is no consensus. At present, it is unclear whether these reasoning biases share common underpinnings or are independent [33].

Based on the previous results of several studies about schizophrenia research [17, 32, 34], our team decided to study the jumping to conclusion bias, bias against disconfirmatory evidence (BADE) and bias against confirmatory evidence (BACE) in patients with schizophrenia and healthy population with high and low score in schizotypy [13] using the Pictures Decision Task [13, 23]. Following the dimensional theory of schizophrenia, we thought that it was interesting to study the population with high schizotypy since it would be useful to understand the etiological mechanisms that there are under schizophrenia spectrum disorders. This understanding could do progress to the own prevention or the early detection of these disorders [35]. The schizotypy is found within the normal variation of population general. Individuals with high schizotypy have a similar psychopathology and cognitive styles than patients with schizophrenia, that is, they are similar qualitatively but they are not similar quantitatively to patients with schizophrenia [13]. For that, we recruited a total of 45 partici‐ pants divided in three groups: 15 patients with schizophrenia and 30 healthy participants (15 high schizotypy and 15 low schizotypy). To measure schizotypy, we used the Community Assessment of Psychic Experiences (CAPE) [36]. Moreover, there are no significant differences between them in age, education, gender, or premorbid intelligence. Once participants were tested, they performed the Pictures Decision Task.

The results of the experiment demonstrated that the patients with schizophrenia displayed jumping to conclusions more easily than control groups (high and low schizotypy). Also, patients with schizophrenia showed confirmatory bias, so that, they were more reticent to change their hypothesis even though it would have new disconfirmatory evidence. Moreover, patients with schizophrenia were less sensitive to the feedback, so they did less use of feedback. For example, in the cue condition, they did not profit in possible solutions when the other two control groups (high and low schizotypy) did it. Therefore, we suggested that jumping to conclusions bias may be related to propensity to hold strong beliefs (high plausibility rating at first stages) and/or to low feedback sensitivity (FS), above all, when the task or context is more ambiguous and difficult (uncued trials). This is corroborated by the fact that all groups reproduced jumping to conclusions in the cued condition, but not in the uncued condition (difficult task), where the patients with schizophrenia and high schizotypy group reproduced the bias more early than low schizotypy group.

Based on these results, we could conclude that jumping to conclusions bias was a general bias because this bias is not only presented in schizophrenia but also in nonclinical population (high‐ and low‐schizotypy healthy populations). However, patients with schizophrenia would show it earlier and stronger than healthy population. It could say that jumping to conclusions bias would be found in a straight line where the patients with schizophrenia would have a greater tendency to show it, followed by populations with high schizotypy and low schizotypy. Furthermore, this line would be influenced by context (more or less ambiguity). Hence, using a controlled or heuristic processing would depend on context and type of participant. Also, we observed that feedback sensitivity could be a factor that affects this bias. However, there is no relationship between jumping to conclusions bias and the two other biases [bias against disconfirmatory evidence (BADE) and bias against confirmatory evidence (BACE)].

To conclude, these results can have an important implication since they could help in the treatment, prevention, or recovery from schizophrenia. For example, the therapy X could try to teach how does more effective hypothesis testing through making a better use of feedback. The jumping to conclusions bias and the other two biases could represent an important therapeutic target. Individual differences in JTC performance could be useful in determining the best course of treatment. Training programs that aim to ameliorate the jumping to conclu‐ sions response style might prove to be an important adjunct to established therapies [37]. An example of this, we can find it with the metacognitive training/therapy [19]. This therapy works on different aspects between the cognitive biases like jumping to conclusions or confirmatory bias. In this therapy there are sixteen modules, which must be done by the patient with schizophrenia. For example, to work the confirmation bias, module 3 is used where patients are informed and explained about this bias and then performed different tasks. A typical task to work this bias would consist in a task with a series of three pictures which are shown in reversed order. The sequences of pictures gradually reveal an ambiguous plot. For each picture, participants are asked to rate the plausibility of four different interpretations. The goal of this task is that patients learn to look for more information before making a judgment and therefore avoiding the confirmation or disconfirmation bias [17, 37].

#### **2.2. Need for closure bias**

between them in age, education, gender, or premorbid intelligence. Once participants were

The results of the experiment demonstrated that the patients with schizophrenia displayed jumping to conclusions more easily than control groups (high and low schizotypy). Also, patients with schizophrenia showed confirmatory bias, so that, they were more reticent to change their hypothesis even though it would have new disconfirmatory evidence. Moreover, patients with schizophrenia were less sensitive to the feedback, so they did less use of feedback. For example, in the cue condition, they did not profit in possible solutions when the other two control groups (high and low schizotypy) did it. Therefore, we suggested that jumping to conclusions bias may be related to propensity to hold strong beliefs (high plausibility rating at first stages) and/or to low feedback sensitivity (FS), above all, when the task or context is more ambiguous and difficult (uncued trials). This is corroborated by the fact that all groups reproduced jumping to conclusions in the cued condition, but not in the uncued condition (difficult task), where the patients with schizophrenia and high schizotypy group reproduced

Based on these results, we could conclude that jumping to conclusions bias was a general bias because this bias is not only presented in schizophrenia but also in nonclinical population (high‐ and low‐schizotypy healthy populations). However, patients with schizophrenia would show it earlier and stronger than healthy population. It could say that jumping to conclusions bias would be found in a straight line where the patients with schizophrenia would have a greater tendency to show it, followed by populations with high schizotypy and low schizotypy. Furthermore, this line would be influenced by context (more or less ambiguity). Hence, using a controlled or heuristic processing would depend on context and type of participant. Also, we observed that feedback sensitivity could be a factor that affects this bias. However, there is no relationship between jumping to conclusions bias and the two other biases [bias against

disconfirmatory evidence (BADE) and bias against confirmatory evidence (BACE)].

To conclude, these results can have an important implication since they could help in the treatment, prevention, or recovery from schizophrenia. For example, the therapy X could try to teach how does more effective hypothesis testing through making a better use of feedback. The jumping to conclusions bias and the other two biases could represent an important therapeutic target. Individual differences in JTC performance could be useful in determining the best course of treatment. Training programs that aim to ameliorate the jumping to conclu‐ sions response style might prove to be an important adjunct to established therapies [37]. An example of this, we can find it with the metacognitive training/therapy [19]. This therapy works on different aspects between the cognitive biases like jumping to conclusions or confirmatory bias. In this therapy there are sixteen modules, which must be done by the patient with schizophrenia. For example, to work the confirmation bias, module 3 is used where patients are informed and explained about this bias and then performed different tasks. A typical task to work this bias would consist in a task with a series of three pictures which are shown in reversed order. The sequences of pictures gradually reveal an ambiguous plot. For each picture, participants are asked to rate the plausibility of four different interpretations. The goal of this

tested, they performed the Pictures Decision Task.

100 Schizophrenia Treatment - The New Facets

the bias more early than low schizotypy group.

Another cognitive bias that has been linked to schizophrenia disorder is the need for closure bias [21, 38]. However, as discussed below this bias has not been studied so deeply as the other biases (jumping to conclusions, BACE and BADE). Therefore, its influence in schizophrenia disorder is not well known yet. Moreover, there are not many studies on its bidirectional relationship with the other implicated cognitive biases in the schizophrenia.

According to Kruglanski [39], the cognitive bias called need for closure is:

"the need to reach a fast decision to have an answer and to escape the feeling of doubt and uncertainty and "freeze" by failing to update".

In addition, this author later adds that this bias could be displayed through the desire for predictability and preference for order and structure and discomfort with ambiguity (need for closure [40]). In general, different studies have evidenced that people who show the need for closure bias have a great need for cognitive closure. They dislike uncertainty and prefer to reach conclusions quickly and with certainty [39–42].

According to Federico et al. [41]:

"They seek to accomplish this goal by "seizing" quickly on any available infor‐ mation to reach conclusions and by "freezing" on these conclusions once they are reached".

Focusing in the study of this bias in schizophrenia, Colbert and Peters [38] demonstrated that members of the general population that are delusion prone had a higher score on the need for closure scale (NCS). Moreover, these individuals displayed jumping to conclusions bias. So they concluded that as the data‐gathering reasoning bias was found in delusion‐prone individuals, this suggests that it may be involved in the formation, rather than merely the maintenance, of delusional beliefs.

In other study, McKay et al. [21] found this bias in patients with schizophrenia. On the other hand, the results of this investigation showed that need for closure and jumping to conclusions biases would not seem related with each other. The intolerance of ambiguity correlated positively with delusion proneness and decisiveness correlated negatively. According to these authors, the delusion‐prone individuals would be more indecisive in everyday life. In addition, the need for closure has been associated with jumping to conclusions since the intolerance to ambiguity contexts would lead to jumping to conclusions [21]. The following year, in 2007, these authors realized other experiments studying the relationship between need for closure and schizophrenia [43]. They wanted to replicate the study of Bentall and Swarbrick [44].

Bentall and Swarbrick thought that patients with delusions may be highly intolerant of ambiguity, that is, they show need for closure and in point the fact that the results of their study confirmed their hypothesis. They found that patients with delusions were highly intolerant of ambiguity and had a higher score in the need for closure scale (NCS). Based on this studio, McKay et al. [43] hypothesized that 22 patients with a history of persecutory delusions would exhibit higher need for closure and a more extreme jumping to conclusions bias than 19 healthy control participants. For that, the participants must realize a probabilistic task and fill out depression and need for closure scale. The results demonstrated that patients with persecutory delusions had a higher score than healthy control group. Therefore, the results support an association between persecutory delusions and need for closure. In addition, they did not find relationship between jumping to conclusions and need for closure.

Within the relationships between cognitive biases (e.g., jumping to conclusions, bias against disconfirmatory evidence, need for closure), we can find the interesting study of Moritz et al. [45]. These authors studied a total of 56 patients with schizophrenia through four independent components: jumping to conclusions, personalizing attributional style, inflexibility and low self‐esteem. The study lends tentative support for the claim that candidate cognitive mecha‐ nisms for delusions only partially overlap, so these mechanisms must be more widely studied in order to have a higher knowledge. Meantime, these authors propose that these biases should be treated independently via behavioral cognitive therapies, which work these biases.

Analyzing all these studies, we see clearly that it is necessary to study more deeply the need for closure and its implication in schizophrenia. Nowadays, it is not known how this bias works in schizophrenia, that is, it is not clear what is its real involvement in the onset, maintenance and relapse of schizophrenia disorder. In fact, if we try to find studies, we will encounter that there are few or almost no one study which attempts to discern what is its influence on schizophrenia. In addition, there are few studies that have tried to investigate the relationship of the implicated cognitive biases in schizophrenia. If the relationship between them is independent, dependent or partially dependent is unclear. In general, it seems that these biases are independent but it is necessary to study better. What is clear is that these cognitive biases play an important role in schizophrenia because they have been found in the same population. In conclusion, the study of need for closure bias and its influence on schizophrenia is needed. In addition, the relationship between these biases in order to obtain better effective therapies should be also examined. While this is not achieved as it is said by Moritz et al. [45], these cognitive biases should be treated independently in the therapies.

#### **2.3. Overconfidence bias**

Finally, another cognitive bias that has been related to schizophrenia disorder is overconfi‐ dence bias [46] because the patients with schizophrenia displayed overconfidence in their choices or interpretations [13, 46, 47]. The overconfidence bias is the tendency to overestimate or exaggerate our own ability [48]. The response confidence is usually enhanced for erroneous judgments in patients with schizophrenia in comparison with healthy controls [13, 14]. In general, the overconfidence bias has been obtained across memory tasks [46, 47]. For example, Peters et al. [49] did an investigation with 27 patients with schizophrenia and 24 healthy controls where they were administrated a developed emotional video paradigm with 5 videos differing in emotionality (positive, 2 negative, neutral and delusional related). After each video, the participants had to do a recognition task. Also, they are asked to say the confidence response, that is, participants must make old‐new discriminations along with confidence ratings. The objective was to see the memory accuracy and meta‐memory deficits. The results demonstrated that in the positive video the patients recognized fewer correct items than healthy controls. The patients with schizophrenia exhibited more high‐confident responses for misses and false memories. So the overconfidence bias displayed by them would be related to higher probability of committing error judgments.

Also, this cognitive bias has been observed in social cognition tasks [50]. Köther et al. [50] carried out a study with 76 patients with schizophrenia or schizoaffective disorder and 30 healthy control participants. In this study, the participants must fill out the Reading the Mind in the Eyes test (Eyes test). Moreover, they had to complement a rating scale requesting response confidence. The results showed that patients with schizophrenia had more high‐ confidence error and fewer high‐confidence correct responses. Besides, this was most clear in patients with formal thought disorder. Therefore, this study supports the implication of this cognitive bias in schizophrenia and its spectrum disorders.

Other interesting study was realized by Moritz et al. [14]. In this study, the authors analyzed the perceptual judgments in patients with schizophrenia. For that, a total of 55 patients with schizophrenia, 58 patients with obsessive‐compulsive and 45 healthy controls participated. These participants had to judge whether the pictures depicted an object or not and how confident they were in this judgment. The results showed that patients with schizophrenia had more overconfidence in their error response and enhanced knowledge corruption index in comparison with healthy controls. However, accuracy score did not differ between patients with schizophrenia and obsessive‐compulsive.

In the study discussed in the previous section [13] also, it was found that patients with schizophrenia showed greater confidence in the early stages of the Pictures to Decision task. In addition, they obtained more errors than healthy controls but the difference was not statistically significant. However, the overconfidence bias is a possible explanation of higher production of errors in patients with schizophrenia.

With respect to the relationship between this bias and the other implicated cognitive biases in schizophrenia, there is not a strong conclusion. In fact, as we saw in the previous cases, it is not clear if the relationship is dependent, independent, or partially dependent. Therefore, its study is necessary for higher understanding, if it is clear that this bias would be involved in the maintenance of schizophrenia disorders. Taking together the exposed results, we could intuit the important implications of these cognitive biases in the onset, maintenance and relapse or recovery in schizophrenia disorders. For that, several new therapies have been created to work in avoiding and recognizing them.
