Verified by 
I have been studying the families-of-origin of patients with borderline personality disorder (BPD) since the late 1970s. As readers of this blog already know, I discovered that the biggest and most common emotional conflict in the parents who produce offspring with the disorder is a conflict over the very act of becoming and being a parent.
As I described in a previous post, the parents think that being great parents and taking care of children is the end all and be all of human existence, but secretly they wish for a variety of different reasons, that they could be spending most of their time doing something else entirely.
Because they are conflicted about their parental roles, they constantly give off mixed messages to their children about what is expected from them. This leaves the children in a quandary about how to meet their parents’ contradictory needs to be both needed by their children and, simultaneously, for them to just go away. The children solve this dilemma by adopting various versions of the spoiler role.
Children growing up with mixed messages from parents is, in fact, a hallmark of families that produce children with any and all personality disorders or repetitive self-destructive behavior. Problematic mixed messages are usually in regard to only certain behaviors or role functions—perhaps just one or—while the parents respond to other child behavior in a coherent, non-contradictory manner.
For those readers who are now adults, how does your family stack up in the continuing double messages department? My colleague Hillel Abramson and I designed an instrument to try to measure the amount and type of double messages adults perceived as receiving from their primary attachment figures, called the Family Interaction Scale or FIS. While designed with only borderline personality disorder (BPD) in mind, it can be used more generally, and you can administer it to yourself to see what areas are problematic for your families.
But first, just a little background.
In order to see what is going on in families, one has to appreciate that conversations between family members have subtexts. A subtext is an underlying, implicit, and often distinct theme in a conversation that is not perceivable to someone who is only paying attention to the literal meanings of the sentences spoken. In order to understand a subtext, one also has to listen to an entire conversation before hidden meanings might become evident. Just a few sentences will not do.
Furthermore, an understanding of the entire history of the relationship between the people having the conversation, not just the current interactions, is also required to tease out important latent themes.
Therefore, appreciating the subtexts of conversations between intimates involves thousands of variables that will be completely different for each dyad or group being evaluated. To think that subtexts of this complexity can be “measured” by standardized questions and answers on a psychological test is ludicrous. For these reasons, truly “empirical” studies are impossible.
In today's world, however, one must try. Some people insist that observations of subtexts are just “anecdotal.” They can therefore just dismiss subtext as an object of scientific study. To get around this just a wee bit, we chose nine of the most common themes in the families of those with BPD that I had observed over the years, and we asked four different groups to rate their parents’ ongoing, current communication in these nine “domains.”
The four groups were: 1) patients with the disorder, 2) those who were subthreshold for the disorder (that is, who met three or four of the criteria for BPD rather than the required five), 3) non-psychotic patients being treated in our clinic who did not have BPD, and 4) normal controls who exibited no psychiatric disorders at all.
The domains are:
1. Praise and criticism
2. Comments to friends and relatives about the subject
3. Reactions to advice given by the parents
4. Blame for the parents’ problems
5. Frequency of advice giving by the parents
6. Degree of parent’s preoccupation with the subject
7. Relationship of a parent’s mood to the subject’s mood
8. Parental reactions to the subject’s work failures and successes
9. Parental reactions to the subject’s romantic failures and successes
For each domain, we provided six to seven options. Because no human relationships are completely consistent all the time, we wanted to avoid saying anything that implied that the parents were “always” one way or another. So we asked subjects to please check only the one item in each domain rating that "came the closest to describing the parent's current behavior towards you most of the time."
For each domain, there was one response pattern that we deemed appropriate. Another response pattern we originally deemed “neglectful.” However, in thinking about it as the results came in, it seemed to us that it would be more accurate to say that these items reflected the fact that the issue in the domain did not seem to matter at all to the parents.
There were two response patterns that we deemed polarized—the parent acted most of the time at one extreme or the other without much in between. There were also two response patterns that were indicative of the pattern in which we were most interested: double messages. The parents either were perceived as blatantly contradictory in their behavior, or they said one thing but seemed to mean something that was contradictory to what they said.
In reading the items below, you should be able to easily figure out which item corresponds to which response pattern.
Subjects in the original research were asked to choose and rate up to three attachment figures from their childhoods who were still alive, still interacting with the subject in the present, and not senile. “Still in contact" could mean in person, by phone, by letter, or even by word of mouth.
These figures had to have served in the parental role when the subjects were between the ages of 5 and 15. We provided the following list:
1. Natural Mother
2. Natural Father
3. Adoptive Father
4. Adoptive Mother
5. First Stepfather
6. Second Stepfather
7. First Stepmother
8. Second Stepmother
9. Father's Father (only if he lived with you)
10. Father's Mother (only if she lived with you)
11. Mother's Father (only if he lived with you)
12. Mother's Mother (only if she lived with you)
13. Brother at least 5 years older than you
14. Sister at least 5 years older than you
15. Other (Specify)
Here were the test items:
___
DOMAIN 1
1. This person consistently praises me no matter what I decide about how I run my life
2. This person consistently criticizes me no matter what I decide about how I run my life.
3. This person consistently goes back and forth between praising and criticizing me for the same decisions.
4. This person is consistently fair in praising and criticizing me for my decisions.
5. This person rarely praises or criticizes my decisions about how I run my life.
6. Regarding my decisions, this person consistently says one thing but really means another.
__
DOMAIN 2
1. This person consistently brags about me to their friends and relatives.
2. This person consistently bad mouths me to their friends and relatives.
3. This person says contradictory things about me to friends and relatives.
4. This person is consistently fair in making remarks about me to friends and relatives.
5. This person consistently does not talk about me with their friends and relatives.
6. This person consistently says one thing about me to friends and relatives but really means another.
__
DOMAIN 3
1. This person acts happy only if I follow their advice.
2. This person acts happy only if I ignore their advice.
3. This person consistently acts unhappy with my reaction to their advice no matter if I follow or ignore it.
4. This person accepts my reaction to their advice even if I disagree with it.
5. This person consistently does not react to how I respond to their advice.
6. Regarding my reactions to their advice, this person consistently acts one way but really feels another.
__
DOMAIN 4
1. This person consistently blames themself for their own problems even when they are not at fault.
2. This person consistently blames me for their own problems even when I'm not at fault.
3. This person consistently goes back and forth between blaming me and themself for their problems.
4. This person is fair in assigning blame for their problems.
5. This person consistently blames others rather than me or themselves for their problems.
6. This person rarely blames anyone for their own problems.
7. In assigning blame for their problems, this person consistently says one thing but means another.
__
DOMAIN 5
1. This person almost never tells me what to do, even if I ask for their advice.
2. This person almost always tells me what to do.
3. This person consistently goes back and forth between telling me to make my own decisions and telling me what to do.
4. This person usually tells me to make my own independent decisions but will offer advice if asked.
5. This person is unconcerned with who makes the decisions in my life.
6. Regarding telling me what to do, this person consistently says one thing but really means another.
__
DOMAIN 6
1. This person consistently expresses worries about how I am doing even when they know I'm doing well.
2. This person never expresses worries about me even when they know I'm in trouble.
3. In expressing worries about how I'm doing, this person goes back and forth between expressing worries and saying nothing.
4. This person is consistently reasonable when expressing concerns about how I'm doing.
5. This person doesn't even bother to find out how I am doing.
6. In regards to worrying about how I'm doing, this person consistently says one thing but really means another.
__
DOMAIN 7
1. This person only acts happy when I am happy.
2. This person only acts happy when I am miserable.
3. This person goes back and forth between acting happy when I'm happy and acting happy when I'm miserable.
4. This person shows sympathy with how I feel but their mood does not depend on how I am feeling.
5. This person almost never reacts to how I am feeling.
6. When I'm happy or unhappy, this person consistently acts one way but really feels another.
__
DOMAIN 8
1. This person does not act happy unless I am succeeding in my career.
2. This person usually acts unhappy about my career accomplishments and relishes in my failures.
3. This person goes back and forth between acting happy about my career successes and relishing in my failures.
4. This person expresses satisfaction with my career successes and sympathizes with my career failures.
5. This person almost never reacts to my career successes and failures.
6. When reacting to my career successes and failures, this person consistently acts one way but really feels another.
__
DOMAIN 9
1. This person does not act happy unless I am having a successful romantic relationship.
2. This person usually acts unhappy about my successes in romantic relationships and relishes in my failures.
3. This person goes back and forth between happy about my romantic successes and relishing in my failures.
4. This person expresses satisfaction with my romantic successes and sympathizes with my romantic failures.
5. This person consistently does not react to how I am doing in romantic relationships.
6. When reacting to my love life, this person consistently acts one way but really feels another.
__
The results of the study were published in 2005 in the psychiatric journal Comprehensive Psychiatry, volume 46, pages 340-352, and they were highly significant and in line with most of our predictions. The fact that they were so significant was even more impressive considering the fact that the number of subjects in each subgroup was relatively small. Discovering small differences usually requires larger samples, so this means that the differences were probably quite large.
We found there were significant differences between groups in regard to appropriate and contradictory responses for the first attachment figure rated. BPD patients endorsed fewer appropriate responses than did the patient controls (p=.001) and than did the normal controls (P=.011). There were no other significant differences among groups regarding appropriate responses for the first rated individual.
BPD patients reported significantly more contradictory responses than did patient controls (p=.004) and normal controls (p=.003). In addition, subthreshold subjects reported more contradictory responses than did patient controls (p=.009) and normal controls (p=.005).
Contrary to prediction, there were no significant differences in the mean number of polarized responses among any of the groups. The polarized responses were indicative of something called splitting: the tendency of people to think of others as all good or all bad. BPD patients are thought to be unable to see shades of grey, but in this study, they were no more likely to endorse polarized response patterns than anyone else. In other words, this was evidence that patients with BPD are quite able to avoid splitting (as I opined in the linked post), at least when thinking about these issues. This was also shown in another way described below.
There were also no significant differences in what we first called neglect responses among any of the groups. This was a bit of surprise because neglect is often present in the parenting styles of those with BPD. However, even when prominent, it tends to oscillate with overinvolvement. But if these items measured lack of interest rather than neglect per se as described earlier, the fact that the parents of subjects with BPD were not characterized as uninterested should have been no surprise. That is because—clinically—even when parents of future BPD patients are neglectful, they tend to blame it on the child!
When chronic unhappiness (dysthymia) was controlled for in comparisons of subject groups, the pattern of significant effects remained the same.
The number of subjects rating two parental figures (N=59) was considerably smaller than the number of subjects who rated at least one (N=93). No significant overall effect between the groups on the way they responded to the FIS regarding person B was found—possibly because of the much smaller sample size. However, the trends were the same.
Normal controls reported significantly more appropriate responses than did BPDs (p=.002), subthresholds (p=.004), and patient controls (p=.024). BPD patients reported significantly more contradictory responses than did normal controls (p=.002) and trending toward reporting more than did patient controls (p=.064), while subthreshold subjects reported significantly more than did normal controls (p=.038). There was no significance between group differences in polarized or neglect responses.
Every contradictory item on the questionnaire was endorsed for person A by at least two of the BPDs (5 percent) with a range of 5 percent to 32.5 percent. The most commonly endorsed contradictory item was, “This person consistently goes back and forth between praising and criticizing me for the same decisions.”
When we examined the FIS by domains, the most commonly endorsed domain in which either contradictory response was endorsed by BPDs for person A was domain 1, with 40 percent. The corresponding percentages were domain 2, 27.5 percent; domain 3, 25 percent; domain 4, 20 percent; domain 5, 32.5 percent; domain 6, 22.5 percent; domain 7, 20 percent; domain 8, 27.5 percent; and domain 9, 20 percent.
The fact that the pattern in the various domains was all over the place within the BPD sample also indicates that each item was considered separately by the subjects and there was no pattern of items being rated globally in one consistent way. Such global ratings might be expected in "splitting" subjects who were unable to look simultaneously at both good and bad characteristics in an individual.
Furthermore, although the mean number of contradictory responses of BPD subjects within the varius domains was considerably higher than the mean for patient controls for person A, it was still quite small—only 2.24 (range 0 to 8 out of a possible 9). The mean for the normal controls, however, was only 0.46. In other words, the normal controls perceived almost NO double messages from their parents.
So readers, go ahead and use the FIS to rate your own families on how often they drive you to distraction with double messages. You of course do not have to share the results with anyone unless you want to.
