OPINION
LAGUEUX, District Judge.This is an appeal from a “final decision” of the Secretary of Health and Human Services denying plaintiff Social Security disability benefits. The matter was referred to a Magistrate for a recommendation on the proper disposition of the matter. The Magistrate by Memorandum and Recommendation issued on August 29, 1986, found substantial evidence on the record to support the Secretary’s decision that plaintiff was not “disabled” within the meaning of 42 U.S.C. § 416(i) and § 423(a). Plaintiff duly objected thereto. Briefs were filed by both parties and the Court heard oral arguments in this matter on November 17, 1986. This case is now in order for decision. Due to the necessity of weighing all the evidence on the record in order to render a proper decision, it is necessary to detail both the lengthy factual and procedural history surrounding this case.
On March 19, 1980, plaintiff was treated at the Roger Williams Hospital’s emergency room following a work related back injury (Tr. 125-127). The medical report indicated that plaintiff had left side paraspinal muscle spasms but that straight leg raising *532was negative (Tr. 126). An x-ray of the spine revealed a narrowing of the disc space between L4 and L5 (Tr. 127); however, the pedicles were intact; the prevertebral soft tissues were unremarkable; and there was not any fracture or dislocation (Tr. 127). Plaintiff was diagnosed as having a low back strain, was treated conservatively, and was discharged in satisfactory condition (Tr. 126).
In July of 1980, plaintiff was examined by M. Howard Triedman, M.D., a board certified neurologist. In a report dated July 15, 1980, Dr. Triedman stated that his examination revealed positive straight leg raising at 45 degrees bilaterally but that there was normal strength in both arms and legs (Tr. 128). Dr. Triedman's findings also indicated slightly decreased knee and ankle jerks (Tr. 128). An electromyogram showed evidence of a mild to moderate neuropathic process affecting what appeared to be the SI nerve roots bilaterally, with minor and poorly defined changes present in the L5 distribution (Tr. 128). Dr. Triedman concluded that his findings suggested the presence of an L5-S1 disc lesion (Tr. 129).
On November 18, 1980, plaintiff was admitted to St. Joseph’s Hospital, complaining of pain in the back and legs (Tr. ISO-138). On examination, plaintiff experienced back pain on extremes of motion beyond about 65 per cent and could straight leg raise on the left to 65 degrees and on right to 80 degrees (Tr. 132). There was no atrophy in the lower extremities and his pulses were palpable (Tr. 132). X-rays of the lumbrosacral spine showed “very mild” disc space narrowing at L4-L5 but no other abnormalities were detected (Tr. 134). A lumbar myelogram revealed “very slight” central bulging at L4-L5; however, there was no evidence of a disc herniation (Tr. 135). Following the myelogram, plaintiff was treated for headaches which the attending physician indicated probably resulted from the myelogram (Tr. 138).
Plaintiff was next examined by David M. Barry, M.D., a board certified neurologist, on January 8, 1981 (Tr. 139, 167). In a report dated January 9, 1981, Dr. Barry noted that plaintiff was still experiencing back pain radiating into the left leg (Tr. 139), but indicated that plaintiff’s pain was not aggravated by coughing or sneezing (Tr. 139), and that there was no tilt, sciatic nerve tenderness, motor weakness or abnormal sensory functions (Tr. 139). Additional findings, however, did suggest that plaintiff did have some lumbar paravertebral muscle spasm and tenderness; plaintiff’s back motions were all restricted to a good 50% of the normal range; straight leg raising was positive at 60 degrees on the left; and that there was a definite decrease in the left ankle reflex (Tr. 139). Dr. Barry concluded that despite the previous negative myelogram, he believed that plaintiff had a ruptured L5-S1 disc, and that as a last resort, surgical exploration might have to be considered (Tr. 139).
On May 11, 1981, plaintiff was hospitalized for back and leg pain (Tr. 140-150). Dr. Joseph Izzi, a board certified orthopedic surgeon, reported that on examination, plaintiff experienced pain, spasms, tenderness, and 80% restricted motion (Tr. 142, 170). Examination of the lower extremities revealed left straight leg raising to 65 degrees, right straight leg raising to 85 degrees and a slight decrease in right ankle jerk. His concluding impression was that plaintiff had a herniated disc at lumbar five, sacral one on the left (Tr. 142). His conclusion, however, remained unconfirmed as of May 12, 1981, when an x-ray of plaintiff’s lumbosacral spine revealed no disc space narrowing or other pathology (Tr. 148). Throughout this time-period plaintiff continued to undergo conservative treatment which included the use of nerve blocks, heat, exercise, etc. Plaintiff was discharged from the hospital on May 16, 1981, in a somewhat improved condition (Tr. 140-150).
In a series of progress notes between April of 1980 and August of 1981, Dr. Izzi reported that plaintiff continued to experience back and leg pain and was being treated conservatively (Tr. 153-157). Dr. Izzi *533indicated that plaintiffs condition was showing no improvement and consistently concluded that plaintiff was “totally disabled” (Tr. 153-157).
Sometime in mid-1981, plaintiff was referred to the Pain Program of Southern New England where he was examined by its medical director, Toussaint A. Leclercq, M.D., a board certified neurosurgeon (Tr. 158, 171). In a report dated June 18, 1981, Dr. Leclercq stated that the neurological examination revealed straight leg raising to be positive for pain at about 20 degrees on the left and 50 degrees on the right (Tr. 158). He also noted that the deep tendon reflexes were present and symmetrical, except for the left ankle jerk which was depressed but not abolished. Dr. Leclercq concluded that plaintiff appeared to have a radiculopathy secondary to a herniated intervertebral disc (Tr. 158). On August 12, 1981, Dr. Leclercq reported that both he and a medical panel had reviewed plaintiffs myelogram and concluded that it revealed “a small indentation at 5-1 on the left side compatible with a very lateral disc” (Tr. 159-160). Dr. Leclercq suggested that conservative treatment should be attempted before surgery was considered, and that, therefore, the panel had recommended a trial of chiropractic manipulation (Tr. 159-160). In a letter dated November 25, 1981, Dr. Leclercq reported that plaintiff had been discharged from chiropractic care because of lack of sustained improvement (Tr. 161). He further reported that the panel believed plaintiff had a herniated disc at L5-S1 and surgery was needed at that time. Dr. Leclercq stated that because plaintiff did not want surgery, the only approach they could offer him was gravity traction and supportive therapy to deal with the pain (Tr. 161).
In August 1982, plaintiff was examined by Stanley J. Stutz, M.D., a board certified orthopedic surgeon (Tr. 165-166, 173). After noting that plaintiff had been in a recent automobile accident (Tr. 166), Dr. Stutz reported that plaintiff had indicated that his back had not changed at all, and that he was experiencing intermittent numbness down his right arm. Plaintiff was taking aspirin in order to relieve these symptoms (Tr. 166). In addition, Dr. Stutz reported marked paraspinal tenderness, spasm, and tenderness over the sciatic notch on the left. There was positive straight leg raising at 30 degrees bilateral with left buttock and thigh pain, as well as a positive cross straight leg raising, a decreased sensation plantar aspect of the left foot and weakness of the toe flexors and extensors (Tr. 166). Reflexes and Babinski response, however, were normal; both leg lengths and the circumference of the calves were equal; and there was no clonus (Tr. 166). At the time, Dr. Stutz’s impression was that plaintiff had a L5 neuropathy, and concluded that plaintiff was “totally disabled” (Tr. 166). Dr. Stutz stated that he would consider anti-inflammatory medication, weight reduction, and surgery (Tr. 166).
In a letter dated March 14, 1983, Dr. Leclercq reported that plaintiff was discharged from the pain clinic because he had failed to follow the recommendation of the panel members that he have surgery (Tr. 178). Dr. Leclercq noted again that plaintiff’s previous myelogram had revealed an indentation at 5-1 on the left side compatible with an irradiculopathy (Tr. 178). On September 26,1983, however, Dr. Leclercq reported that plaintiff had returned because of a severe exacerbation of pain which radiated into the right lower extremity (Tr. 186). On examination, plaintiff had severe muscle spasm and a decreased right ankle jerk. Dr. Leclercq recommended a CT scan and acupuncture treatment (Tr. 186).
In May 1983, plaintiff underwent a neurological examination by Barry Levin, M.D. (Tr. 179-184). In a neurology evaluation report dated May 13, 1983, Dr. Levin reported that on examination, plaintiff was mildly obese and in no acute distress (Tr. 180). Plaintiff’s back showed tenderness to palpatation and percussion in the lumbosacral region, in the midline, and laterally. There was also moderate paraspinal muscle spasm bilaterally in the lumbosacral region and decreased curve reversal. Plaintiff *534noted marked pain with straight leg raising at 20 degrees on the left and 30 degrees on the right. There was similar pain with flexion of the knee at the abdomen (with or without extension) and with the Patrick maneuver (Tr. 180). Dr. Levin stated that nearly every movement of the left lower extremity elicited pain, and this pain exceeded that obtained on the right side (Tr. 180). Plaintiffs gait was described as marked, favoring the left lower extremity. Reflexes were traced throughout, although with reinforcement, the lower extremities were 1-2 + with questionable slight decrease of the left hamstring jerk (Tr. 180).
With respect to plaintiffs daily activities, plaintiff indicated that these included doing dishes, taking short walks, and driving a car (Tr. 182). In addition, plaintiff reported that 47% of his day was spent standing, 40% lying down, and 13% sitting down (Tr. 182) . Dr. Levin noted that plaintiffs description of his daily activities was consistent with the objective findings (Tr. 182). Dr. Levin completed a physical capacities evaluation which indicated that the plaintiff could do the following: sit for two hours at a time, stand and walk, each for one-half hour at a time; sit and stand, each for a total of four hours during an eight hour day; walk for a total of two hours during an eight hour day; occasionally lift up to five pounds; use both hands for repetitive grasping, pushing and pulling of arm controls, and fine manipulation; and use the right, but not the left foot for repetitive pushing and pulling of leg controls (Tr. 183) . Dr. Levin indicated that plaintiff was not able to bend, squat, crawl, climb, or reach (Tr. 183). In concluding his report, Dr. Levin stated that previous studies were suggestive of active disc disease. (Tr. 181). He noted, however, that a differential diagnosis, in view of sensory changes, would have included other abnormality at T5-6 such as a extramedullary lesion (Tr. 181). Dr. Levin concluded that plaintiff “appears to be totally disabled on the basis of his back pain,” and recommended that plaintiff be further reviewed to rule out a T5-6 extramedullary lesion (Tr. 181).
In June 1981, plaintiff underwent a psychological evaluation by Gerald D. Fontaine, Ph.D., as part of his participation in the pain clinic (Tr. 162). In a report dated June 24, 1981, Dr. Fontaine noted that plaintiff displayed some depressive symptoms along with generalized anxiety (Tr. 162) . Dr. Fontaine concluded, however, that he did “not see [plaintiff] as a good candidate for psychotherapy other than supportive measures which can help us alleviate his pain” (Tr. 162).
Plaintiff was subsequently evaluated in July of 1981 by Robert S. Carson, M.D., a board certified psychiatrist (Tr. 151-152, 168). In a report dated July 2, 1981, Dr. Carson noted that plaintiff was vigorously opposed to surgery and had stopped taking medication because he did not like the side effects (Tr. 151). While Dr. Carson stated that plaintiff “seems to be in moderately severe pain to the point where he is currently disabled” (Tr. 152), he noted that no pyriformis spasm was elicited (Tr. 152). With respect to plaintiffs mental condition, Dr. Carson found no evidence of a severe psychiatric problem and described plaintiff as being “well adjusted emotionally” (Tr. 152). Finally, as a result of plaintiffs vigorous opposition to surgery, he recommended conservative therapy such as acupuncture or chiropractic manipulation (Tr. 152), with a long range recommendation of vocational rehabilitation because it was improbable that plaintiff could continue in heavy construction work (Tr. 152).
In a follow up visit on January 6, 1982, Dr. Fontaine reported that plaintiff appeared to be in “a fairly good frame of mind showing no depression and very little anxiety” (Tr. 163). He noted that plaintiff was fearful of an operation and did not intend to have one in the future, but was handling his problem “without undue anxiety” and was making “good progress” (Tr. 163) . On February 10, 1982, Dr. Fontaine reported that plaintiff was experiencing a considerable amount of pain and had been drinking up to a six pack of beer a day to alleviate the pain (Tr. 163). Dr. Fontaine noted that plaintiff recognized his condition was not improving but plaintiff denied that *535he was severely in need of an operation in order to improve his condition (Tr. 163). On March 31, 1982, Dr. Fontaine reported that plaintiff had begun physical therapy utilizing gravity traction, which gave him up to two hours pain relief twice per day (Tr. 164). Dr. Fontaine stated that plaintiff was experiencing some anxiety because his insurance company indicated that he could return to light work (Tr. 164). Plaintiff disagreed because he felt he was unable to stand on his feet for any length of time (Tr. 164). Dr. Fontaine recommended that plaintiff follow through with the Vocational Rehabilitation Department “where hopefully he can learn a new skill through a training program they can provide” (Tr. 164).
While plaintiff was undergoing treatment for his alleged disability in early 1982, he filed an application for a period of disability insurance benefits claiming inability to work since the initial date of his injury (Tr. 77-80). Plaintiff’s application was denied on March 26, 1982 (Tr. 85), after which, plaintiff sought reconsideration of his claim by the Social Security Administration (Tr. 86). Here too, plaintiff’s claim for disability benefits was denied (Tr. 95). A hearing de novo was then held before an Administrative Law Judge (AU) and on March 1, 1984, the AU held that between March 19, 1980, and November of 1981, the degree of pain and severity of discomfort suffered by the plaintiff appeared not to be of sufficient severity so as to prevent him from engaging in a full range of sedentary work (Tr. 32). After November of 1981, however, the AU concluded that the severity of plaintiff’s symptoms, including the degree of pain and impairment, indicated that the plaintiff was not capable of a full range of sedentary work activity (Tr. 33). With respect to this latter period, the AU went on to find that plaintiff was still not “disabled” because he had failed, without good reason, to follow the prescribed treatment of his treating physicians (back surgery) which would be expected to restore plaintiff’s ability to perform sedentary work (Tr. 36).
After the AU’s opinion was rendered, plaintiff requested review of the decision before the Appeals Council. By letter dated August 14, 1984, the Appeals Council granted plaintiff’s request (Tr. 187-88), and on December 31, 1984, rendered an opinion denying the plaintiff disability benefits on the grounds that plaintiff was not “disabled” given his relatively young age (forty-one), his high school education, and present ability to engage in sedentary work (Tr. 15).
Following the Appeals Council’s decision, plaintiff brought this action to review the “final decision” of the Secretary of Health and Human Services. The matter, then, was referred to the Magistrate who recommended that the Court uphold the Secretary’s decision.
In order to do so, however, this Court must resolve the following three issues: (1) Whether there exists substantial evidence on the record to indicate that plaintiff possessed a residual functional capacity for engaging in sedentary work between March 19,1980 and November of 1981? (2) Whether there exists substantial evidence on the record to indicate that plaintiff possessed a residual functional capacity for engaging in sedentary work between December 1, 1981 and June 30, 1982? (3) If the substantial evidence test is not met with respect to either time-period, whether plaintiff’s refusal to follow “recommendations” on the part of his treating surgeons bars him from collecting benefits for the periods involved?
In considering the first and the second of these issues, it is useful to note that while there is no dispute as to the time-period over which plaintiff is claiming disability benefits (March 19, 1980 to June 30, 1982), the precise point in time that plaintiff became disabled (if at all) remains an area of contention. Regarding this question, the Appeals Council found that plaintiff was not “disabled” over the entire period between March 19, 1980 and June 30, 1982. The AU, however, determined that while plaintiff was not “disabled” between March 19, 1980 and November of 1981, plaintiff was “disabled” from December 1, 1981 *536through June of 1982. While the standard of review that must be applied is the same under either approach, for the sake of clarity, this Court chooses to apply the substantial evidence test to the entire record within the time-frame adopted by the AU.
In addition to this time-frame consideration, this Court notes that in determining whether the Secretary’s decision is supported by substantial evidence on the record, it is bound by the Social Security Administration’s regulations defining the term “disability.” Pertinent to the consideration of this case are those regulations which provide that one must first assess the claimant’s residual functional capacity (RFC) in order to determine whether the claimant possesses a capacity for doing his previous work in spite of his impairment. 20 C.F.R. § 404.1545. This is done by assessing a claimant’s physical abilities, mental impairments and “other impairments” 20 C.F.R. § 404.1545. Since it is undisputed that the claimant does not possess any mental impairment or “other impairment,” such as epilepsy or impairment of the senses, these two criteria need not be considered any further with respect to making a determination in the present case.
Having determined that plaintiff’s RFC prevents him from engaging in his previous or a similar occupation, it is then necessary to consider the claimant’s age, education, and work experience along with his RFC in order to determine what other kind of work the claimant may be able to do in spite of his impairment in order to arrive at a disability decision. 20 C.F.R. § 404.1561. This last step is reached by two alternative methods. First, one applies a precisely matching medical-vocational guideline pattern to determine whether or not the claimant is disabled 20 C.F.R. § 404.1569. Second, if the guideline pattern does not precisely match the age, education, work experience and RFC of the claimant, then consideration must be given to all relevant facts in accordance with the definition of each criterion in the regulations in order to reach a disability decision. 20 C.F.R. § 404.1569.
From this discussion of the guidelines, it is evident that their application is conditioned upon arriving at a precise determination of the four criteria listed above. While there appears to be no dispute concerning the plaintiff’s age, education or work experience, a dispute does remain concerning the degree of the plaintiff’s RFC. Thus, an examination of the evidence in light of the definition of RFC serves as the only basis for this Court’s inquiry with regard to the first two issues.
As explained above, the only criterion relevant to determining the plaintiff’s RFC in this case is “physical abilities.” After assessing this factor, one is then required to determine whether it allows the claimant to accomplish a particular degree of work. Here, the degree of work it is alleged that plaintiff is capable of doing is that form known as “sedentary work.” This type of work is defined by the regulations in the following manner:
(a) Sedentary work. Sedentary work involves lifting no more than 10 pounds at a time and occasionally lifting or carrying articles like docket files, ledgers, and small tools. Although a sedentary job is defined as one which involves sitting, a certain amount of walking and standing is often necessary in carrying out job duties. Jobs are sedentary if walking and standing are required occasionally and other sedentary criteria are met.
20 C.F.R. § 404.1567. Keeping these definitions in mind, as well as the time-frame established by the AU, this Court will proceed to determine whether there exists substantial evidence on the record to indicate that plaintiff possessed a residual functional capacity for sedentary work between March 19, 1980 and November of 1981.
In assessing the plaintiff’s “physical abilities” between March 19, 1980 and November of 1981, it is necessary to examine the plaintiff’s medical records to determine the extent of his impairment 20 C.F.R. § 404. 1545. For the first few weeks following plaintiff’s injury the record reflects that plaintiff was in some discomfort (Tr. 126).
*537This is substantiated by the x-ray of plaintiffs spine which revealed a narrowing between the L4 and L5 disc space (Tr. 127), and the paraspinal muscle spasms felt by the plaintiff at this time (Tr. 126). This evidence, however, is inconclusive as to the extent of plaintiffs injury as exemplified by the fact that straight-leg raising was negative (Tr. 126) and that plaintiff was diagnosed as having only a “low back strain” for which heat and bed-rest were recommended (Tr. 126).
Nor is there any evidence in the record to suggest that the plaintiff, during this time-period, was incapable of fulfilling the definition of sedentary work. The one reference in the plaintiff’s Emergency Room record, which might be probative on this point — “unable to sit” (Tr. 126) — is inconclusive since it does not indicate the period over which plaintiff was unable to do so, or make any reference to plaintiff’s lifting ability. Clearly plaintiff could sit for some period of time since he could do so for a period of four hours during an eight hour day several months later after his condition had deteriorated (Tr. 183).
A like conclusion can be drawn from the medical data on record from August of 1980 through November of 1980. While the plaintiff’s electromyogram taken in July of 1980 was “abnormal” so that one doctor concluded that it suggested the presence of an L5-S1 disc lesion (Tr. 129), x-rays taken of the lumbro-sacral spine in November of 1980 continued to show only a “very mild” disc space narrowing at L4-L5 (Tr. 134). In addition, while leg raising had deteriorated somewhat since March of 1980, plaintiff was still capable of obtaining 65 degrees on the left and close to 80 degrees on the right (Tr. 132). Finally, a myelogram, taken at this time, revealed only a “very slight” central bulging at L4-L5 and showed no evidence of disc herniation (Tr. 135).
Between December of 1980 and November of 1981, the evidence shows signs of deterioration in plaintiff’s condition. For example, during this time-period, there was a definite decrease in ankle jerks as opposed to only the “slight decrease suggested earlier in 1980 (Tr. 139), and plaintiff’s back motion became increasingly restricted from 50% of normal in January to 80% in May of 1981 (Tr. 142). In addition, positive results of a straight leg raising test increased markedly to 20 degrees on the left and 50 degrees on the right as compared to 65 degrees on the left and 85 degrees on the right in April of that year (Tr. 155, 158). Finally, plaintiff’s deteriorating condition during these months is most marked by the diagnoses of three physicians who examined the plaintiff. Dr. David M. Barry, a board certified neurologist, concluded that despite the previous negative myelogram, he believed that the plaintiff had a “ruptured” L5-S1 disc (Tr. 139, 167). Dr. Joseph Izzi, a board certified orthopedic surgeon, reached a similar conclusion in May of 1981 (Tr. 142), and also consistently concluded that the plaintiff was “totally disabled” (Tr. 153-157). Lastly, Dr. Toussaint A. Leclercq, a specialist in neurosurgery, concluded, after examining plaintiff, that plaintiff appeared to have “radiculopathy secondary to a herniated interverbral disc” (Tr. 158). This conclusion was corroborated by a review of plaintiff’s myelogram which indicated a “small indentation at 5-1 on the left side compatible with a very lateral disc” (Tr. 159).
Despite these findings indicating that plaintiff possessed a painful back injury throughout the early and middle parts of 1981, there is still not enough evidence present so as to undermine the Secretary’s conclusion because there is no evidence on the record of a connection between plaintiff’s injury and his mobility to engage in sedentary work. Indeed, despite the three physicians’ conclusions, there is some evidence that plaintiff’s condition remained constant or improved slightly during this time. For instance, in January of 1981, the record indicates that plaintiff was experiencing symptoms similar to those he had been experiencing since he was first injured: back pain radiating into his lower leg and leg raising of approximately 60 degrees on the left (Tr. 139). Throughout this time-period, plaintiff’s physicians con*538sistently recommended that plaintiff undergo only conservative treatment rather than more extensive treatment such as surgery (Tr. 140, 155, 159, 160). Indeed, after undergoing such treatment, plaintiffs condition appeared to improve somewhat (Tr. 139, 140). Most importantly, however, this Court relies on the fact finder’s ability to appraise the credibility of the plaintiff’s testimony regarding his disability, See, Universal Camera Corp. v. N.L.R.B., 340 U.S. 474, 496, 71 S.Ct. 456, 468, 95 L.Ed. 456 (1951). In this respect the AU concluded the following:
In determining the claimant’s residual functional capacity, the Administrative Law Judge has carefully considered the claimant’s allegations of severe, disabling lower back pain and radiating left leg pain and the degree of impairment the claimant described. Although the Administrative Law Judge realizes that the claimant experienced some degree of pain and discomfort, it was not of sufficient severity to persuade him that the claimant was incapable of engaging in a full range of at least sedentary work activity from March of 1980 through November of 1981. (Tr. 32).
This assessment by the AU, when weighed with the indications of some improvement in the plaintiff’s condition and the fact that only conservative treatment was prescribed during this time, presents enough evidence so as to substantially support the finding by the Secretary for this time-period. In so concluding, that is not to say that the evidence does not support a finding that the plaintiff was in some degree of pain or limited in his daily functions. Rather, it only means, that there does exist substantial evidence on the record so as to support a finding of the Secretary that plaintiff was capable of engaging in sedentary work for this period.
Having resolved the first issue in favor of the Secretary, it is necessary to examine the evidence with regards to the second issue: whether there exists substantial evidence on the record to indicate that plaintiff possessed a residual functional capacity for engaging in sedentary work between December 1, 1981 and June 30, 1982?
By the end of November of 1981, it had become clear that plaintiff indeed had suffered a herniated disc at L-5, S-l and that surgery was needed at this time (Tr. 161). While plaintiff continued to undergo conservative treatment after November, it was not because such treatment was “recommended” by his physicians, but rather because plaintiff was fearful of having back surgery (Tr. 164). The conservative treatment’s primary function, thus, was to alleviate the pain felt by plaintiff (Tr. 163). More importantly, the post-November time-period is marked by evidence which links the degree of plaintiff’s incapacity to his alleged inability to engage in sedentary work. For example, during this time-period, plaintiff used a cane in order to walk; standing and climbing was “painful;” lying down relieved this pain; and plaintiff started to experience a previously unreported numbness and “pins and needles” radiating along with the pain into his legs (Tr. 72, 165).
At plaintiff’s hearing, the degree of plaintiff’s incapacity was further described. According to plaintiff's testimony, plaintiff was able to sit for approximately one-half hour before he became uncomfortable (Tr. 69). Out of a thirty day month plaintiff indicated that he would spend ten days just resting in bed all day; he would spend another ten days experiencing severe pain that allowed some walking about; and he would spend the remaining ten days experiencing less severe pain that enabled him to engage in more substantial activities (Tr. 69, 70). Finally, plaintiff indicated that prior to his accident he used to play basketball and football, coach little league and jog (Tr. 70). As of early 1983, however, he participated in none of these activities (Tr. 71). Although plaintiff’s own testimony regarding the limitations resulting from his impairment may be subject to little credence in itself, plaintiffs testimony in this case is supported by objective medical evidence. A physical examination of plaintiff in May of 1983, revealed that plaintiff’s *539pain was worse with prolonged sitting, standing or walking and was brought on by any bending or lifting (Tr. 179). Furthermore, plaintiff was taking six Darvocet (a powerful pain killer) per day in addition to the aspirin which he had previously been taking to relieve his pain (Tr. 179). It is also notable that plaintiffs straight-leg raising performance had substantially deteriorated so that by this time he felt pain at 20 degrees on the left and 30 degrees on the right (Tr. 180) instead of the previous 60 and 80 degree marks.
Finally, during this examination, Dr. Barry Levin, a specialist in neurology, made the following findings regarding plaintiffs physical abilities. At one time, plaintiff could sit for two hours, stand for one-half hour, and walk for one-half hour (Tr. 183). During an entire eight hour day, Dr. Levin reported that plaintiff could sit for four hours; stand for four hours; and walk for two hours. With respect to plaintiffs carrying ability, he found that plaintiff could “occasionally” (defined as one percent to thirty-three percent of an eight hour day) carry up to five pounds but never more than ten pounds (Tr. 183). Lastly, Dr. Levin found that plaintiff was completely unable to “bend,” “squat,” “crawl,” “climb” or “reach” (Tr. 183).
Comparing these findings to the regulations’ definition of sedentary work clearly indicates that after November of 1981, plaintiff simply did not possess a residual functional capacity to engage in sedentary work. Not only was plaintiff unable to meet the ten-pound requirement of that definition (the plaintiff could only occasionally lift five pounds) but he failed to meet the regulations’ “sitting” requirement. Clearly, if sedentary work requires “occasional” standing, defined at most as thirty three percent of an eight hour day, (two hours and forty minutes), then such work requires a claimant to sit for the remainder of an eight hour period (five hours and twenty minutes). Plaintiff, however, could sit at most for four hours during an eight hour day, and thus, fails to meet the requirements of sedentary work by one hour and twenty minutes. In practice, plaintiff probably failed the sitting standard by an even greater amount given his inability to sit for more than one-half hour at a time without discomfort.
As if these findings were not clear enough regarding plaintiff’s inability to perform sedentary work, the ALJ himself found the following:
By November of 1981, the medical evidence of record indicates that the claimant’s condition had apparently deteriorated to the extent that surgery was recommended. Based on significant findings on physical examination and the degree of pain the claimant alleged, both consultative evaluators, Dr. Stutz and Dr. Levin, concluded that the claimant was disabled at the time of their examinations. It is concluded that the severity of the symptoms, including that of alleged severe pain and the degree of impairment the claimant described would result in a severely limited residual functional capacity for less than a full range of sedentary work activity.
(Tr. 33). As has been previously noted, the AU’s assessment of plaintiff’s credibility regarding his pain and degree of impairment must be taken into account. Universal Camera Corp., 340 U.S. at 496, 71 S.Ct. at 468. When this is done, such evidence supports the medical evidence on record which indicates that plaintiff did not possess any RFC for sedentary work after November of 1981.
Having reached this conclusion, it is necessary to determine if the decision of the Secretary for this period may still be supported by substantial evidence on the record. In assessing the plaintiff’s RFC, the Secretary concluded “for the reasons discussed above in combination with the minimal findings observed on x-ray and myelogram, the Appeals Council has determined ... [that plaintiff possesses] an ability to perform a full range of sedentary work” (Tr. 14, 15). Although it is unclear from the decision as to what other “reasons,” in addition to the x-ray and myelogram the Secretary was referring to, three *540reasons become apparent after careful scrutiny of the text of the Secretary’s decision. First, the examinations at one time or another revealed normal motor and sensory functions and reflex activity. Secondly, the claimant’s back pain was apparently not aggravated by either coughing or sneezing. Thirdly, no atrophy was present since the initial diagnosis of plaintiff’s back problem.
With respect to all three of these “reasons,” it should be noted that there is absolutely no expert opinion on the record linking them to a finding regarding the plaintiff’s physical ability for engaging in sedentary work. As a result, it is apparent that the Secretary either makes such a connection through her own leap of logic or by assuming the mantle of an expert herself. For example, there is no indication in the Secretary’s decision as to how normal sensory and reflex functions aid plaintiff in engaging in sedentary work. More disturbing are the conclusions drawn by the Secretary from the second and third reasons. From the second, the Secretary concludes that because plaintiff’s pain was not aggravated by coughing or sneezing, it was unlikely that plaintiff had a “herniated disc with nerve root impingement.” From the third, the Secretary concludes that the lack of atrophy (or reporting thereof) somehow militates against the existence of plaintiff’s back problem. The problem with these conclusions is that in reaching them, the Secretary is not merely weighing inconsistent data and selecting that which it finds more probative. See, 20 C.F.R. § 404.1527. Rather, the Secretary is assuming the role of an expert, interpreting certain symptoms of plaintiff, and reaching a medical conclusion. See, Rosado v. Secretary of Health & Human Services, 807 F.2d 292, 293 (1st Cir.1986). This the Secretary may not do. Id. When viewed from this perspective it becomes immediately apparent that the other “reasons” do not support the Secretary’s decision regarding plaintiff’s RFC.
As for the x-ray and myelogram, these facts also do not support the conclusion reached by the Secretary. The Secretary concluded that the x-ray and myelogram contributed to a finding of RFC because they showed only a “mild” narrowing of the L4-5 disc space and only a “very slight” central protrusion at L5-S1 respectively. While this may be some evidence of RFC, its value as substantial evidence of such a conclusion is lessened for two reasons. First, as indicated above, there is no expert testimony connecting the ambiguous x-ray and myelogram results with plaintiff’s capacity to work. Secondly, as the Secretary, herself, concedes, as of 1981, the evidence was reinterpreted by plaintiff’s treating neurosurgeon so as to indicate a herniated disc (Tr. 14). At best then what can be said of the x-ray and myelogram is that they are inconclusive regarding plaintiff’s RFC for sedentary work, and thus, require other evidence to support a finding that the plaintiff is able to engage in sedentary work. This other evidence is completely lacking in the present case.
Finally, one need only examine the Secretary’s analysis of the exertional requirements in order to discover that the record is devoid of further evidence supporting the Secretary’s findings. Although the Secretary was “cognizant” of the assessment of an examining physician that the claimant could lift only five pounds (and also presumably of the fact that the claimant could only sit for a total of four hours out of an eight hour day), the Secretary still concluded that the claimant could lift up to ten pounds and could sit and stand for a total of eight hours during a standard working day (Tr. 15). As is evident from the previous discussion regarding these factors, the record suggests plaintiff simply could not lift the amount or sit for the length of time ascribed to him by the Secretary. It is clear, then, that there is no substantial evidence on the record to indicate that the plaintiff possessed a RFC for sedentary work between December 1, 1981 and June 30, 1982.
Having decided the second issue in favor of plaintiff, it becomes necessary to decide whether plaintiff’s refusal to follow “rec*541ommendations” for surgery on the part of his treating physicians bars him from collecting benefits for the time-period December of 1981 through June of 1982.
In deciding this issue, this Court (as well as the Secretary) is guided by the First Circuit’s decision in Schena v. Secretary of Health & Human Services, 635 F.2d 15 (1st Cir.1980). In that case the Court of Appeals announced that the following four-part test must be complied with before the Secretary may terminate disability benefits under 20 C.F.R. § 404.1530.
(1) The impairment must have been amenable to treatment ... [that could be expected] to restore the claimant’s ability to work.
(2) The treatment must have been prescribed.
(3) The treatment must have been refused.
(4) The refusal must have been willful; wilfulness [sic] does not exist where there is a justifiable excuse for the refusal.
Schena, 635 F.2d at 19.
Let us apply this test to the facts of case. It is readily apparent that the third part of the test has been satisfied. Plaintiff clearly refused to have back surgery “recommended” by Dr. Leclercq at the end of November of 1981 (Tr. 161, 162). Furthermore, there is no dispute that plaintiff continued to refuse this surgery through June of 1982 (Tr. 179).
It is far less clear, however, whether the first, second, and fourth parts of the test have been satisfied. Regarding the first part of the test, the First Circuit indicated that the mere suggestion that surgery would improve “the claimant’s condition” was not enough evidence so as to sustain a positive finding that the treatment in question would restore claimant’s ability to work. Schena, 635 F.2d at 19. In the present case, the record is devoid of any facts indicating that surgery would enable the plaintiff to engage in sedentary work. Nor is it clear whether plaintiff’s treatment was “prescribed.” With respect to this portion of the Schena test, the First Circuit defined “prescribed treatment” in the following manner: “that various physicians suggested the operation does not necessarily mean that they prescribed it.” Schena, 635 F.2d at 19. The Court then cited evidence from the record that Schena’s physician had not “urged Schena to undergo treatment he was fundamentally opposed to and had no objection to the patient continuing physical therapy for a relatively indefinite period of time.” Id. While the distinction between a “prescription” and a “mere recommendation” can be termed semantic at best, it appears that in order to amount to a “prescription”, the treatment in question must have been “urged” upon the patient and the physician’s “prescription” must have been more than “tentative.” Benedict v. Heckler, 593 F.Supp. 755, 760 (E.D.N.Y.1984) (dictum).
In the plaintiff’s case, by November of 1981, plaintiff was informed that “surgery was needed” (Tr. 161), and in fact, he was discharged from chiropractic care due to lack of sustained improvement (Tr. 161). Thus, plaintiff’s physicians were no longer “tentative” in their conclusions as to whether he should undergo surgery. Rather, surgery was the form of treatment urged upon plaintiff, and conservative treatment became a secondary form necessary only because plaintiff was afraid of the former. Thus, one can conclude in the present case that by the end of November of 1981, back surgery was the “prescribed” treatment for plaintiff.
Finally, it is necessary to consider whether there was a justifiable excuse for the plaintiff’s refusal to undergo back surgery. In making such a determination the First Circuit has indicated that one must consider the following three factors:
(1) The likelihood of success as provided by percentage figures.
(2) Whether alternative treatment is available.
(3) The pain and danger of the contemplated surgery.
Schena, 635 F.2d at 19.
With regard to the first of these factors, it is apparent that none of the physicians of *542record have provided percentages indicating the likelihood of success were plaintiff to undergo back surgery. As to the second factor, it is clear that alternatives such as heat, bed rest, acupuncture, chiropractic manipulations and gravity traction were available to help plaintiff cope with his condition (Tr. 161). Furthermore, plaintiff was willing to try these measures and obtained some relief from them (Tr. 67, 163, 164).
Finally, with respect to the third factor, while it is unclear from the record as to the precise surgical operation prescribed for plaintiff by his physicians, it is clear that any such surgery would be “major” (Tr. 65), and would involve at least some degree of risk and pain to the patient. Schena, 635 F.2d at 20 (quoting Ratliff v. Celebrezze, 338 F.2d 978, 981 (6th Cir. 1964)) (“It is common knowledge that spinal surgery is often dangerous and entails much pain and suffering”). Weighing this factor, along with the first two, one is led inescapably to the conclusion that plaintiff was acting reasonably in refusing to undergo the treatment prescribed by his physicians. In terms of the Schena test, then, plaintiffs reasonable fear of surgery constituted a “justifiable excuse” for refusing such treatment so as to render such refusal “not willful”. Given that both this part, and the first part of the Schena test are not met in the present case, this Court concludes that plaintiff cannot be barred from receiving disability benefits from December of 1981 through June of 1982 as a result of his refusal to follow the prescribed treatment by his treating physicians.
Based upon the conclusions reached by this Court on the three issues discussed above, the decision of the Secretary is affirmed in part and reversed in part, and the case is remanded to the Secretary with instructions to award plaintiff disability benefits for the period of December 1981 through June of 1982.
It is so Ordered.