Introduction

^{Bowlby (1969/1998)} stated that one of the fundamental conditions for the development of secure bonds in children is the presence of a sensitive figure who responds to their signals and fosters a harmonious interaction. In this sense, sensitivity is defined as the caregiver’s ability to perceive and identify the communication implicit in the child’s behavior, interpret it properly, and respond to it with ownership and promptness (^{Ainsworth, 1969}). Ainsworth’s (¹⁹⁶⁷, ¹⁹⁶⁹) findings made it possible to verify that sensitivity precedes the formation of attachment bonds and is directly connected to the level of security that the child develops (^{Ainsworth et al., 1978/2015}).

In the origins of attachment theory, the mother was considered the leading caregiver figure within the family nucleus; therefore, sensitive behavior towards children was expected from her. However, evidence has shown that other people may also show sensitivity in the care of children, such as the father (e.g., ^{Brown, Mangelsdorf, & Neff, 2012}; ^{Hallers-Haalboom et al., 2014}; ^{Romero, 2007}), grandparents (e.g., ^{Crowling, Seeman, & Göpfert, 2015}; ^{Howes & Spieker, 2016}), other relatives (e.g., ^{Cassidy, 2016}), and even those in charge of daily care (e.g., ^{Carbonell, 2013}; ^{Salinas-Quiroz et al., 2014}; ^{Salinas-Quiroz, 2015}).The essential idea is that these people focus their attention on the child’s physical and emotional needs, see things from his/her point of view, synchronize their activities with those of the child, and adjust to his/her emotional states, developmental period, and particularities (^{Howes & Spieker, 2016}; ^{Morales, Cruz-Martínez, Posada, & Carbonell, 2014)}.

Since children’s needs evolve as they grow, a sensitive caregiver will care for children according to their level of development. Thus, although the typical preschooler does not require the permanent physical presence of the caregiver (^{Bowlby, 1969/1998}), it is essential that this person is accessible and available, responds to communication, stimulates him/her to explore the environment while monitoring his/her activities (^{Kobak, Zajac, & Madsen, 2016}; ^{Waters & Cummings, 2000}), makes agreements with him/her (^{Salinas-Quiroz et al., 2014}), anticipates situations involving conflict, and sets limits on behavior (^{Posada, Kaloustian, Richmond, & Moreno, 2007}).

Based on these considerations that refer to early childhood, in 1998, Posada, Moreno, and Richmond created the Maternal Behavior for Preschoolers Q-set (MBPQS) to assess the sensitivity of caregivers of preschool children (^{Posada et al., 2007}). The MBPQS comprises 90 items that describe normative behaviors of sensitive care and whose construction included behaviors observed or reported by mothers of preschoolers, as well as some references from the Maternal Behavior Q-set (MBQS) by ^{Pederson and Moran (1995}), which evaluates the care of infants from 1 to 2 years old, and the Parental Secure Base Support and Supervision Q-set by ^{Waters, Gao, and Elliot (1998/2003)}, which determines secure base support in stages after infancy. For each of the 90 items, there is a score called the “sensitivity criterion” that was prepared with the participation of professionals in developmental psychology (^{Posada et al., 2007}).

Fifty-five MBPQS items constitute four scales that represent specific dimensions of sensitive behavior. These scales are (^{Posada et al., 2007}): (1) Contribution to harmonious interactions (CHI), consisting of 20 items (e.g., item 3: “Participate in games with the child, for example, play in the sand, run with him/her”); (2) Secure base-support (SBS), with 22 items (e.g., item 44: “When the child shows the caregiver something s/he is playing with, the caregiver asks about it, makes positive comments, and encourages the child to do something with it”); (3) Supervision and monitoring (SUP), with 8 items (e.g., item 61: “The caregiver seems to be aware of the child even if s/he is not in the same room”); (4) Setting limits (SL), with 5 items (e.g., item 67: “When the caregiver establishes rules and prohibitions for the child in an activity, s/he explains the reasons). Contemporary studies have reported Cronbach’s alpha values for the MBPQS scales ranging between .90 (CHI) and .52 (SL) in Colombia; between .96 (CHI) and .62 (SL) in Peru; between .95 (CHI) and .62 (SUP) in Mexican immigrants (^{Posada et al., 2016}); and between .96 (CHI) and .71 (SUP) in Ecuador (^{Díaz, Andrade, Espinosa, Nóblega, & Núñez del Prado, 2018}).

Regarding its use, the MBPQS has been applied in a self-reporting manner to describe an ideal, optimally sensitive maternal figure (^{Posada et al., 2007}); it has also been used observationally to evaluate the sensitivity of the caregiver figure towards the child during their habitual interactions in their natural environment (^{Nóblega, 2012}; ^{Posada et al., 2007}; ^{Posada et al., 2016}). To respond to the items, the Q-sort methodology is used, which consists of the ranking of the 90 item statements written on individual cards. First, three groups are created. The first group corresponds to the caregiver figure’s characteristic behaviors, the second group to neither characteristic nor uncharacteristic behaviors, and the third group to uncharacteristic behaviors. Each group is then divided into three more, which results in a total of nine groups; in each, the 10 statement cards must be sorted so that group 9 corresponds to the very characteristic behaviors and group 1 corresponds to the very uncharacteristic behaviors. Each item receives a score ranging from 9 to 1 according to the group into which it was sorted (^{Posada et al., 2007}).

The results of the MBPQS can be interpreted on a global level, by scales, or by items (^{Posada et al., 2007}). In the first case, the scores assigned to the caregiver figure are correlated with the scores of the instrument’s sensitivity criterion, from which the sensitivity index is obtained. At the scale level, the obtained scores are contrasted using differences in means with the criterion score of each dimension. Finally, at the item level, the scores are compared using the sensitivity criterion for each item.

Although the MBPQS has been adapted to the Latin American context (^{Posada et al., 2007}), it has also been linguistically adapted to the idiomatic expressions of Peru (^{Nóblega, 2012}), where a group of professionals and mothers developed ideal sensitivity profiles of caregivers of typical preschool children. After comparing these profiles’ scores with the MBPQS sensitivity criterion scores, an average correlation of .83 was found for the professionals’ profiles and .81 for the mothers’ profiles (^{Nóblega, 2012}). Additionally, the MBPQS has been used in university studies conducted in Peru with father figures (e.g., ^{Grández, 2016}; ^{Luna Victoria, 2015}) and with daycare workers (e.g., ^{Espinoza, 2016}).

Although the findings indicate that the MBPQS criteria are adequate to assess the quality of caregivers’ sensitive behavior towards preschool children, when a minor is diagnosed with some type of special condition, the situation changes (^{Karst & Van Hecke, 2012}). Usually, parents need time to adjust to the new situation and to reassess their expectations, at which time they often experience feelings of insecurity about their ability to properly handle the situation, as well as uncertainty regarding the future (^{Gupta, 2007}). Research on the subject reports high levels of stress in parents of children with developmental disorders (e.g., ^{Greeff & Nolting, 2013}; ^{Gupta, 2007}), especially when it comes to the autism spectrum (ASD) (e.g., Karst & Van Hecke, 2012; ^{Keenan, Newman, Gray, & Rinehart, 2016}; ^{Martínez & Bilbao, 2008}; ^{Ruble, Murray, McGrew, Brevoort, Wong, 2018}).

ASD is a neurodevelopmental disorder characterized by persistent deficits in communication and social interaction, as well as repetitive and stereotyped patterns of behavior, interests, and activities. These deficits are manifested on a scale of 1 (light) to 3 (severe) according to the levels of severity and help required (American Psychiatric Association (APA), ²⁰¹⁴, ²⁰¹⁶). Although ASD symptoms are present from birth, they are not always evident in early stages of development, so the diagnosis is usually made after two years of age. Worldwide, approximately 1% of the population is considered to have some form of autism, and that for every girl, there are four boys with the disorder (^{APA, 2014}).

Despite ASD’s characteristics, there is evidence that sensitive behavior by caregivers encourages the development of secure attachment ties in children with autism, which favors their social development (e.g., ^{Capps, Sigman, & Mundy, 1994}; ^{Kahane & El-Tahir, 2015}; ^{Koren-Karie, Oppenheim, Dolev, & Yirmiya, 2009}; ^{Teague, Gray, Tonge, & Newman, 2017}). Therefore, the evaluation of the sensitivity of caregivers of children with ASD is essential insofar as it contributes valuable information for individualized intervention programs (^{Keenan et al., 2016}; ^{Kiani & Nami, 2017}).

In digital databases, four studies carried out until 2018 reported on the evaluation of the sensitivity of caregivers of children with ASD (see ^{Capps et al., 1994}; ^{Koren-Karie et al., 2009}; ^{Pechous, 2001}; ^{van IJzendoorn et al., 2007}). In the case of Capps and his collaborators (1994), the filming of a game session between mothers and children was used as an evaluation tool. In ^{Pechous' study (2001)}, ^{Pederson and Moran's (1995}) MBQS was used. In van IJzendoorn et al.’s (2007) and de ^{Koren-Karie et al.’s (2009)} studies, the Emotional Availability Scale (Biringen, 1998/2008) was used. The use of the MBPQS for mothers of children with autism (N = 12) was conducted using self-reports only in a psychology thesis in Peru, in which an average sensitivity index of .64 was found (^{Chiaravalli, 2011}).

Therefore, given the lack of information on the subject, the objective of the present study was to explore the content validity of the MBPQS for its application with caregivers of preschool children with ASD residing in Ecuador. To this end, the participation of expert judges was sought to evaluate the linguistic adaptation of the instrument, of caregiver figures to develop ideal profiles of caregivers of children with typical development, and of professionals with autism expertise to construct ideally sensitive profiles of caregivers of children with ASD. For the profile constructions, ^{Nóblega’s (2012}) study using the MBPQS in Peru and ^{Mesman and collaborators’ (2016}) study with Pederson and Moran’s MBQS used in several countries were taken as models.

Materials and Methods

Results

Ideal sensitive behavior of caregivers of typical children

Among the average scores for the ratings assigned by the 10 participating caregiver figures to each of the MBPQS items and the scores for the instrument’s sensitivity criterion, a correlation index of .867 was found (95% CI = (.808, .912), p < .001), which indicates high similarity between both rating groups.

Concerning the sensitivity dimensions, between the scores of the criterion by scales and the means of the ideal profiles constructed by the caregiver figures, statistically significant differences were present for CHI and SUP. In the case of CHI, t(9) = 3.58, p = .006, the mean of the scores assigned by the participants (M = 7.38) was higher than that of the criterion (M = 7.10); conversely, for SUP, t(9) = -3.58, p = .006, the mean of the criterion (M = 7.59) was higher than that of the participants’ scores (M = 6.81). For SBS and SL, there were no significant differences. The results indicate that, in contrast to the norm for the sensitivity criterion, the participants of this study consider that an ideal caregiver figure acts as a secure base and establishes behavior limits for the typical child as expected; however, the caregiver contributes more to harmonious interactions with the child and supervises and monitors him/her less.

Regarding behaviors with scores close to the criterion (Table 1), 10 items were identified that showed differences between 0.05 and 0.15 points with respect to the criterion score. Four of these 10 items correspond to the CHI, SBS, and SL scales, while there were no similarities for SUP. The remaining six items are not part of the sensitivity dimensions.

Table 1: Similar sensitive care behaviors between the ideal profile of caregivers of typical children and the sensitivity criterion 

Regarding behaviors with high discrepancies (Table 2), the 10 items that had the highest differences to the MBPQS sensitivity criterion were selected. These differences were between 3.7 and 1.8 points. Five of the 10 selected items are not part of the sensitivity dimensions, while the other five correspond to the SBS, CHI, and SUP scales. The largest number of high discrepancies (items 23, 27, and 54) were found for CHI, while for the SL scale, there were no high discrepancies, which suggests that the participating caregivers expect, in general, that the establishment of children’s limits is similar to the instrument’s norms.

Table 2: Sensitive care behaviors with the most discrepancy between the ideal profile of caregivers of typical children and the sensitivity criterion 

Ideal sensitive behavior of caregivers of children with ASD

The average scores of the ratings assigned by the 10 professionals to the MBPQS items were calculated and correlated with the instrument’s sensitivity criterion; an index of .859 (95% CI = (.789, .907), p < .001) was found, showing high similarity between the two groups of scores. When contrasting the scores of the ideally sensitive profile of caregivers of children with ASD with the ideal profile of caregivers of typical children created by the caregiver figures, an index of .955 was found (95% CI = (.938, .968), p < .001), which indicates a significant and very high correlation between the two groups.

Based on the sensitivity dimensions, a statistically significant difference was found between the MBPQS criterion score and the mean of the ideal profile made by the professionals only for SUP, t(9) = -5.28, p = .001. This result suggests that the professionals participating in this study consider that an ideally sensitive caregiver figure fosters harmonious interactions with the child with ASD, provides secure base support, and establishes limits on his/her behavior according to established norms but also supervises and monitors the child differently than expected. For the ideal profile of caregivers of typical children, a similar result was found in the supervision and monitoring scale.

However, when comparing scales for the ideally sensitive profile of caregivers of children with ASD and those of children with typical development, no statistically significant differences were found (p > .05). This result suggests that, in general, the participants consider that both types of figures contribute to harmonious interactions, provide secure base support, and supervise, monitor, and set limits for the children in a similar way, regardless of whether or not they have autism.

Figure 1 shows an overview of the average scores of the scales for the MBPQS and for the ideally sensitive profiles of caregivers of children with typical development and those with ASD.

Figure 1: Means of the MBPQS sensitivity dimensions and the profiles of the caregivers of typical children and those with ASD 

Regarding the behaviors with scores similar to the criterion (Table 3), 12 items were identified with differences between 0 and 0.1 points. Of these 12 items, seven are not part of the sensitivity dimensions, while the remaining five correspond to the CHI, SL, and SBS scales. The CHI scale had the most items with similar scores to the criterion (items 17, 78, and 85). There were no similarities in the SUP scale, which was also found for the development of ideal profiles of caregivers of typical children. This result demonstrates that, in general, in the Ecuadorian context, the supervision and monitoring of children has similar characteristics for both groups of children, regardless of if they are characterized by typical development or ASD. In addition, three items showed similarity with the criterion in the ideal profile of caregivers of typical children; these items are 17 (CHI scale), and 8 and 41, which are not part of the sensitivity dimensions.

Table 3: Similar sensitive care behaviors between the ideal profile of caregivers of children with ASD and the sensitivity criterion 

Regarding the largest discrepancies between the ideally sensitive profile of caregivers of children with ASD and the MBPQS normative criterion, 10 items were identified with differences in absolute values between 4.5 and 1.85 points (Table 4). Three of these items are not part of the sensitivity dimensions, while the other seven belong to the SBS, SUP, and CHI scales. No such high discrepancies were found for the SL scale, nor were they found in the ideal profile of caregivers of typical children. This result indicates that in setting limits, participants expect caregiver figures to demonstrate behaviors similar to those of the criterion, regardless of whether or not the child has ASD. Additionally, four of the 10 selected items also presented high discrepancies in the construction of ideal profiles of caregivers of typical children.

Table 4: Sensitive care behaviors with the greatest discrepancy between the ideal profile of caregivers of children with ASD and the sensitivity criterion 

Discussion

In response to the current structure and functioning of families both in Ecuador and in the Latin American region, the term “attachment figure” was introduced in the linguistic adaptation of the MBPQS to substitute for “mom” and “mother” that appear in the original instrument so it can be used to assess the sensitivity of the child caregiver figure, whether the mother or not. In Peru, for example, some studies have used the MBPQS to examine the sensitive behavior of father figures (e.g., ^{Grández, 2016}; ^{Luna Victoria, 2015}) and of workers in children centers (e.g., ^{Espinoza, 2016}). ^{Nóblega and Posada (2016}) carried out an analysis of the results obtained in these studies and found that, regardless of the caregiver’s sex, in the Peruvian population studied, there was a relationship between caregivers’ sensitivity and the quality of children’s attachment. In fact, the Observatorio de los Derechos de la Niñez y Adolescencia (Observatory for the Rights of Children and Adolescents -ODNA-, 2010) highlights the interchangeable roles that are present among members of current families due to several factors such as time spent in workplaces, diversity in family types, and migration, among others, which has resulted in childcare being shared between the parents or even delegated to other people (^{Carbonell, 2013}; ODNA, 2010).

Regarding the creation of ideally sensitive profiles of caregivers of typical children by the participants, a high and statistically significant correlation was found between the MBPQS criterion scores, very similar to that found in Peru by ^{Nóblega (2012}). This finding confirms the universality of the sensitivity construct proposed by ^{Ainsworth (1967}, 1969), which has been investigated by several contemporary authors regarding the caregiver-preschool child relationship (e.g., ^{Kobak et al., 2016}; ^{Posada et al., 2007}; ^{Salinas-Quiroz et al., 2014}). With respect to the analysis of the MBPQS scales, the results indicate that, for the participants in this study, an ideal caregiver for the Ecuadorian context functions as a secure base for the typical child and establishes limits for his/her behavior in the same way as stipulated in the instrument; however, in comparison with the MBPQS norms, the caregiver’s contribution to harmonious interactions with the child is greater, while the level of supervision and monitoring is lower.

On the other hand, when comparing the ideal profile of caregivers of typical children with the sensitivity criterion scores by item, high similarities were found for contribution to harmonious interactions, secure base support, and establishment of limits but not for supervision and monitoring; in the latter scale, statistically significant differences were found with the criterion, especially in items that describe sensitivity behaviors or characteristics or non-characteristics. With regard to the high discrepancies, it was found that, compared to the norm, the participating caregiver figures consider that although an ideal caregiver carries out more activities based on what the child pays attention to and responds more quickly to the child’s positive signals, that does not mean that s/he is necessarily willing to agree to the child’s wishes. These findings are consistent with the cultural practices observed in the Ecuadorian context in which the study was conducted, in relation to the level of care that adults usually give to preschool-age children, and which are manifested through the use of diminutives in language, displays of emotion, and playful behaviors, among other forms of interaction and expression.

In reference to the exploration of evidence of the MBPQS’s validity in its application with caregiver figures of preschool children with ASD residing in Ecuador (this study’s objective), positive, significant, and high correlations were found between the ideally sensitive profile developed by expert professionals in autism and the instrument’s normative criteria, as well as with the ideal sensitivity profile constructed by the caregiver figures regarding caregivers of typical children. These results are indicators that the MBPQS is valid in terms of its content for the evaluation of the sensitivity of caregivers of children on the autism spectrum.

Regarding the scales, SUP was the only dimension in which a statistically significant difference was found with the sensitivity criterion. This difference was also present in the ideal profile of caregivers of the typical population, which reflects the presence of cultural practices related to the level and characteristics of child supervision. Indeed, in the context in which the study was conducted, it has been observed that, in general, adults are less aware of children’s activities and allow them to be more self-managing in resolving their conflicts.

On the other hand, when comparing the scores of the scales of the profiles of caregivers of typical children and those of children with ASD, no significant differences were recorded, which indicates that in the studied environment, it is expected and desirable that the two types of caregiver figures act similarly with the children with regard to contributing to harmonious interactions, secure base support, supervision and monitoring, and setting limits.

Regarding the analysis by item, high similarities were found between the ideally sensitive profile of caregivers of children with ASD and the MBPQS sensitivity criterion. In three of these items, the similarity was also present between the ideal profile of caregivers of typical children and the criterion. Additionally, in the same way as with the profile of caregivers of typical children, similarities did not occur with the SUP scale; however, it was observed that discrepancies were present in behaviors that were neither characteristic nor uncharacteristic, which reaffirms previous findings in the sense that in the Ecuadorian context, generally the supervision and monitoring of children tends to differ from what is stated in the criterion, regardless of if the children have ASD. Additionally, the highest discrepancies between the ideal profile of caregivers of children with ASD and the criterion were found in the same items as the ideal profile of caregivers of typical children, which indicates consistency in the considerations made by the participants when developing the prototypical profiles of both types of caregivers.

Conclusion

This study found high correlations between the prototypical profiles developed by the participants and the MBPQS normative sensitivity criterion. Only for the supervision and monitoring dimension were there significant differences between the criterion score and the ideal profiles of caregivers of both typical children and children with autism, which allows for the determination that adult supervision of minors in the Ecuadorian context differs from that proposed theoretically. Additionally, a considerable amount of similar behavior has been identified between the prototypical profiles developed by the participants and the MBPQS sensitivity criterion. The few high discrepancies were found in items from both profiles, which would indicate that the participants, in general, projected a high expectation of adult attention towards children whether or not they have ASD; however, that expectation does not mean that caregivers necessarily agree to the minors’ wishes. Finally, it is important to note that the findings of both profiles reflect high consistency, which suggests that to create the ideal characterizations of caregiver figures, the participants took into account Ecuadorian social and cultural factors.

Consequently, the results of this study confirm that no specific Q-sort criteria are needed to evaluate child caregivers’ sensitivity in the Ecuadorian environment and that the MBPQS is valid in terms of its content for the application with caregiver figures with both children with typical development and with autism. It should be emphasized that the use of this instrument requires prior training by professionals due to the particularities of the Q-sort methodology.

It is important to note that the information on the caregiver-child with ASD relationship collected with the MBPQS is useful and valuable both for clinical practice and for research on autism linkage relationships. As documented in several studies, the sensitive behavior of caregiver figures encourages the development of secure attachment bonds and, therefore, cognitive and socio-affective development (e.g., ^{Fearon & Belsky, 2016}) both in typical children and in children with autism (^{Kahane & El-Tahir, 2015}; ^{Teague et al., 2017}). Therefore, the available information can serve as a basis for the design of intervention and support programs that meet the needs of children with the disorder and their parents (^{Keenan et al., 2016}; ^{Kiani & Nami, 2017}).

Finally, as previously mentioned, it is important to emphasize that the participants in this study previously developed ideal security profiles for typical children and those with ASD using the Attachment Q-set (AQS) (^{Díaz, 2018}), which could have affected the increase of the ideal sensitivity scores. On the other hand, it should be considered that the participating caregivers belonged to medium-high socioeconomic levels and that the professionals were academics with postgraduate training; therefore, the ideal profiles they created possibly corresponded to caregiver figures of children with typical development and with ASD from cities and the middle class. For these reasons, the results of this study should not be extrapolated to groups with different characteristics than those of the population studied.