Department of Psychological Sciences, University of Missouri

Spring, 2022

My Contribution

• Conducted secondary research to understand the role of attachment in parenting

• Coded and analyzed interview data

• Analyzed quantitative data using R

• Presented the findings

Tools used:

• Qualtrics

• IBM SPSS Statistics

• R programming languages

• Microsoft Excel and Powerpoint

Aim

• To understand mothers' expectations for parenting as a function of their own attachment

Background

Attachment

Attachment is a psychological bond that an infant develops with their primary caregiver in the first year of their life based on consistent caregiving patterns they experience (Bowlby, 1969/1982). These attachment-relevant experiences become represented as a cognitive schema called Internal Working Models (IWMs). IWMs of attachment encompass the view of the self, others, and future relationships (Bowlby, 1973; Bretherton 1985, 1991). More importantly, they are thought to guide the processing of social information (Dykas & Cassidy, 2011). In line with theory, attachment variation in adulthood has been linked with differences in the speed at which social information is processed and biases in attention to emotional cues (Dykas & Cassidy, 2011).

Furthermore, research has found that adults perceive attachment-relevant information differently as a function of their attachment (Dykas & Cassidy, 2011). Specifically, secure adults are thought to openly process potentially painful social information because their history of attachment experiences has informed them that this information is tolerable and support will be provided. However, when faced with the same potentially painful social information, insecure adults exclude the information as a defense by either subconsciously diverting their attention away or limiting access to that information (Emmichoven et al., 2003).

EEG - ERP

To collect mothers' neural responses to attachment-relevant information such as caregiving cues, we (Family and Child Development Lab) employed the Electroencephalogram (EEG) methodology. In particular, we used Event-Related Potentials (ERPs) since they have excellent temporal resolution and highlight distinct stages of cognition. For this study, we used the N400 ERP component ("N" = negative, "400" processing onset after 400 milliseconds from stimuli presentation). It is typically used in language-based studies in which semantic expectations are to be violated. Specifically, sentences with unexpected endings exhibited a larger deflection in amplitude than expected sentences (Federmeier & Kutas, 2002). An example of an expected sentence would be “I take my coffee with cream and sugar” vs unexpected would be “I take my coffee with cream and dog” (Kutas and Hillyard, 1980) (Figure. 1 ).

Also, N400 has been proven to be useful in detecting expectancies for attachment-relevant cues as a function of attachment security in a previous study (Zayas et al., 2009).

Lexical Decision Task

The Lexical Decision Task (LDT) was adapted from a previous study (Zayas et al., 2009) and computerized.

Mothers were to instructed to indicate target stimuli as words or nonwords as quickly and as accurately as possible via a button box. The auditory primes were either attachment-relevant or not attachment-relevant and the targets were sensitive words, insensitive words or nonword.

A total of 80 audio primes and 320 target stimuli were used (Figure. 2)

Lexical Decision Task

For each trial, a fixation cross appeared on the screen, and then 200 ms later, an audio prime was played lasting approximately 4 s. Next, after another fixation cross, the target stimuli appeared on the screen for 500 ms (Figure. 3)4.

Mothers completed four 80-trial blocks. In each block, half of the trials involved nonword targets that were either paired with attachment primes or non-attachment primes and the other half were word targets

Attachment Script Assessment

Mothers' attachment was assessed via an interview task called the Attachment Script Assessment (ASA; Waters & Rodrigues-Doolabh, 2004). They were given six 12-word prompts, each with the title of the story (Figure. 4). Using these words, they were asked to tell stories, The two adult stories (“Jane and Bob’s Camping Trip” and “The Accident”) and two child stories (“Baby’s Morning” and “Doctor’s Office”) designed to evaluate attachment representation.

Narratives were coded on a 7-point scale (1 = low, 7 = high) for:

• Secure Base Script Knowledge (SBSK): Attachment-relevant problems are effectively resolved (Waters and Rodrigues-Dooblah; 2004)

• Deactivation: Attachment-relevant themes minimized

• Hyperactivation: Attachment problems heightened

• Anomalous content: Presence of bizarre themes

Motivation

Despite the evidence that one's attachment affects how one processes social information, the majority of research has focused on attachment biases in the processing of general emotional cues. Only a handful of studies have examined the processing of social info specific to parent-child relationships. Some studies have found that parents’ attachment is associated with their neural responses reflective of attention to infants' faces

(see Fraedrich et al., 2010; Groh and Haydon, 2018; & Leyh et. al., 2016).

However, little is known about the contribution of attachment to parents’ expectations regarding attachment relationships with children.

Examination

Given this information, we examined mothers’ N400 amplitudes to insensitive (v. sensitive) responses to infant attachment cues (e.g., crying) as a function of mothers’ attachment representations.

Results

• The results showed that attachment variation was associated with mothers’ neural responding to caregiving cues only in response to attachment primes (Figure. 5)

• Specifically, in response to attachment primes, attachment deactivation was associated with mothers’ N400 amplitude to sensitive v. insensitive caregiving cues (Figure. 6)

• Attachment secure base script knowledge was modestly associated with mothers’ N400 amplitude in the same condition; however, this did not reach conventional standards of significance

Discussion

Findings demonstrated that mothers’ attachment representations were specifically associated with neural responses reflective of caregiving expectations to infant attachment cues. This provides evidence in support of the specificity of the effect of attachment on expectations regarding parent-child relationships.

Findings also indicated that attachment deactivation was associated with a smaller difference in N400 responding to sensitive v. insensitive caregiving to infant attachment cues. This reflects a dampened expectation that infant attachment signals should be responded to sensitively.

Given that deactivating individuals have a caregiving history in which their attachment signals were ignored or rejected, these findings suggest that a history of unresponsive caregiving contributes to dampened expectations which in turn may undermine the quality of care they provide their infant.

Taken together, these findings contribute to the broader literature on attachment and social information processing. Specifically, they advance our understanding of the contribution of attachment to the processing of social information by providing novel evidence which shows that mothers’ attachment representations contribute to expectations specific to parent-child attachment relationships.