What’s the deal with isolation work? The devil is in the details

If you want to grow your biceps as much as possible, should you include curls in your training program or just stick with the compound pulling movements you already do on back day? Many would tend to answer this question with something along the lines of, “of course you should include curls, that’s just common sense. The biceps are involved in pull-ups and rows, but to maximize a muscle’s growth you should do an exercise that maximally uses that muscle.” However, others will argue that you can build impressive arms with only compound pulling/pushing movements, and they actually have a fair amount of scientific support for that claim. But as always, the devil is in the details.

Isolation exercises refer to single-joint (SJ) movements, meaning that movement is occurring only at – you guessed it – a single joint. Technically, no muscle is ever truly isolated (at the very least, there are some muscles acting to stabilize the body/limb), but for practical purposes, these exercises place virtually all of the burden on a single muscle group. Think curls, lateral raises, leg extensions, etc. Compound, or multi-joint (MJ) exercises, involve multiple muscle groups and movement at multiple joints, like a squat, bench press, row, and so on.

Because MJ exercises incorporate multiple muscle groups at the same time, they’re often considered more efficient, and I can’t argue with that. I have no doubt that doing a program consisting of only MJ training will yield a very respectable physique. But the question we’re trying to get at isn’t whether or not bench press will grow your triceps or if deadlifts will grow your hamstrings – it’s whether they will optimally grow those muscles to bring about not just impressive results, but the best results you could possibly get. There’s some, but not a ton, of research on this, and unfortunately a whole lot of limitations involved, but let’s see what we can get from it.

Available Data

To start off, some work has been done examining muscle growth after doing only SJ training versus only MJ training. To look at this, equating volume between groups is super important, as well as keeping other variables (range of motion, tempo, proximity to failure, rest time, etc.) as similar as possible.

One study (12) compared changes in non-bone fat-free mass via DEXA between two groups doing either SJ or MJ training exclusively; the MJ group did 4 sets of 6-8 to failure and the SJ group did 4 sets of 12-18 to failure, equating for volume-load. They found no significant differences between groups, and if anything, a trend favouring the MJ group (average fat-free mass gain of 5.5% versus 3.5%). However, looking at the study design, I don’t agree that volume was actually matched between groups. While both groups did the same number of hard sets of hamstrings, calves, abs, chest, and delts, there were differences resulting in more volume for the MJ group for quads (4 sets each of leg press and squat vs 4 sets of leg extensions), glutes (4 sets each of deadlifts and squats vs… nothing), back (4 sets each of lat pulldowns and seated rows vs 4 sets of pullovers), biceps (4 sets each of lat pulldowns and seated rows vs 4 sets of curls), and triceps (4 sets each of bench press, incline bench press, and military press vs 4 sets of triceps extensions). So overall, I’m not sure this was really a fair comparison, and the measure of whole-body fat-free mass increase isn’t really what we want either; a more nuanced measure would be an improvement. Still, we could interpret this as a win for SJ training, as even with quite a bit less volume for about half of the muscle groups, they experienced statistically equivalent changes in fat-free mass.

As an interesting aside based on this study – people often ask how to “count volume” for muscle groups involved in MJ movements, but that isn’t the prime mover (i.e. the triceps during a bench press). Some of the data from this study would suggest that an appropriate way for counting sets toward a “secondary muscle” would be 1/3-1/2 of that which we would count for the prime mover. For example, every set of lat pulldowns could be counted as 1 set for lats and 1/3-1/2 of a set for biceps. But as we will see, that conclusion isn’t supported super well by other data that suggest no difference in muscle growth with matched volume. For now, it’s merely my best guess based on my interpretation of the scant research, the limitations of many of the measurements taken, and practical experience.

Two other studies have looked at the “only MJ vs only SJ” comparison, with conflicting results. Gentil et al. (7) found that matched volume between lat pulldowns and curls resulted in the same increase in biceps muscle thickness, while Mannarino et al. (9) found a larger increase (about twice as much) in biceps thickness after doing curls compared to dumbbell rows. I tend to prefer the design in the Mannarino study since all subjects did curls with one arm and rows with the other – acting as their own controls really cuts down on the uncertainty from individual variation. However, another important point could be the exercise selection. Gentil used lat pulldowns, which tend to involve a larger range of motion for the biceps than dumbbell rows do. That could explain why biceps growth was similar in that study, while rows, with their shorter ROM, led to inferior growth in the Mannarino study.

Zooming out – we can’t draw strong conclusions with so little data, but with all else being equal, differences in hypertrophy from SJ and MJ training could just come down to which MJ exercise you choose. If you’re super limited on time and have to choose the minimum possible number of exercises, I’d go with MJ movements that take secondary muscles through the largest ROM, like a pulldown or pull-up with a supinated grip rather than a row, or maybe a close grip bench press rather than a wide grip bench press with a big arch.

That’s great, but what about those of us for whom time isn’t the bottleneck? We’re not limited to choosing just one exercise, so would we get a benefit from simply piling some SJ work on top of MJ work? To try to answer this question, we want to make sure groups in these studies have as many variables matched as possible (range of motion, tempo, proximity to failure, rest time, etc.), but now we don’t want to see equated volume. The entire point of adding SJ training on top of MJ training is to increase the volume on those “secondary” muscle groups, so equating volume would actually sabotage their study. Here’s a spoiler alert on the front end: a predominant measure used in these studies is flexed arm circumference, which is exactly what you think it is – not really the precision we’re hoping for with science. Still, the evidence has merit, we just need to be extra cautious and keep in mind that such a rough measurement leaves a large risk of type II errors (not seeing an effect that actually happened).

The first to take a crack at testing whether adding SJ training on top of a MJ program results in more gains was Gentil et al. (technically, an NSCA conference abstract (13) was presented years earlier – it never went through peer review or publication, but the results are right in line with a lot of the subsequent literature on the topic, so make of it what you will). They had untrained subjects split into two groups: one group did bench press and lat pulldowns and the other did the exact same thing, with the addition of triceps extensions and curls, bumping up the volume for the arm muscles. The changes in biceps thickness were the same in both groups (in line with the point I made earlier about ROM in MJ exercises), but flexed arm circumference, which would have included gains in triceps mass as well as biceps, was a tiny bit larger for the SJ+MJ group, although it didn’t quite reach statistical significance (8).

In trained subjects, some data suggest a small but not statistically significant benefit of adding SJ work (6). This is a bit perplexing since we would normally assume that trained lifters would need more volume than untrained lifters, and so adding in SJ work would be beneficial. My explanation is that this study was only 8 weeks long and only used flexed arm circumference to measure growth. Trained lifters will be growing more slowly anyway, so if a difference between groups did exist, it would have been pretty small after only 8 weeks and simply wrapping a tape measure around someone’s arm may not have been sensitive enough to pick it up.

Finally, we recently saw that the triceps grew more when doing both skull-crushers and bench press when compared to doing either one by itself in a study from Brandao et al. (5). Using MRI, the authors discovered that the different heads of the triceps grew more or less depending on the movement. This fits in nicely with some earlier data from Ogasawara and colleagues, who have a couple of studies measuring muscle growth in the chest and triceps following a bench press training program (10,11). While they weren’t actually trying to answer a question about SJ vs MJ training, they found that the chest grew almost twice as much as the triceps did. This could be interpreted as evidence that the bench press, which involves both the triceps and chest, may not be enough to maximize triceps growth. However, we don’t necessarily know that all muscles have an equal propensity for growth, so we can’t use the Ogasawara studies alone to prove any point about triceps growth since no comparisons were made between different training programs. That being said, combining that data with the recent Brandao study and the other research mentioned above, I feel pretty comfortable saying that bench press alone is great for chest growth, but maybe not optimal for triceps.

The integrity of data collected by Barbalho et al. has recently been questioned in a white paper by Vigotsky and colleagues (15). Several studies by Barbalho and his research group have aimed to address the question of whether adding SJ exercise to a MJ program is beneficial (1–4), however, due to the current investigations regarding that data, those studies were omitted from this review.

Zooming out again – I’d say that the evidence leans slightly in favour of including SJ work, but not by much. With longer studies and more precise measurement techniques, I’d hypothesize that a more defined contrast would emerge, but we can’t say that for certain. Another thing you may have noticed is that no muscles other than biceps and triceps have been tested. There’s definitely room for more work to be done in this area, but I still say that adding in SJ work, while accounting for total volume and fatigue, will probably be a benefit with very little risk of downside. If time isn’t an issue and you’re willing to invest more time for diminishing returns, then I say “why not.”

Programming Considerations

I have a feeling if you’re reading this article then you’re like me and are willing to put in some extra work for that little extra something, so let’s explore how we might want to intelligently go about doing that. There are several points that we want to consider when it comes to our exercise selection:

Does the muscle act as a prime mover?

If the muscle never acts as a prime mover during whatever MJ movements you’re doing, I find it unlikely that it will be growing as much as possible. From a purely logical perspective, why would it? For example, if you’re doing barbell rows, what reason would there be for your rear delts to grow as much as they could when your big back muscles are doing most of the work and should, therefore, be a higher priority for increases in strength/size to get better at barbell rows? Similarly, are your rear delts the limiting factor in your barbell row performance? No way! We know that going at least within throwing distance of failure is important for hypertrophy, but at that point, your back is going to be what feels super fatigued, and I’d be willing to bet that your rear delts are still at a pretty low level of local fatigue. Maybe I’m just not doing it right, but I’ve never had to stop a set of barbell rows because my rear delts were burning too bad. In this sort of situation, I’d add in a little bit of work where the rear delts are the prime mover, like some sort of reverse fly, to make sure all bases are covered.

Does the muscle go through a full range of motion?

As discussed briefly above, if a muscle never goes through a full range of motion during MJ training, it might benefit from either choosing a different MJ exercise that will allow a larger ROM or adding in a SJ exercise to target that muscle through its full ROM.

Does the muscle get loaded in a stretched position?

Related to the above comments on range of motion, a muscle most likely benefits from being loaded in a stretched position. If this never occurs during your chosen MJ exercises, having some sort of SJ movement to achieve it may be beneficial. One example of this could be the lateral delts – they’re unlikely to experience any loaded stretch during an overhead press or upright rows, but you might get a little extra something out of a cable lateral raise.

Biomechanical considerations

As shown in the Brandao study mentioned above, the different heads of the triceps may respond best to different exercises (5). Additionally, some muscles cross more than one joint and perform more than one function, like the hamstrings. In such a situation, it probably makes sense to include exercises that utilize those different functions – for hamstrings, that means some sort of hip-hinge and some sort of leg curl, which have been shown to result in different EMG readings at different portions of the hamstrings (14). This doesn’t necessarily indicate differences in hypertrophy, but if you’re geared up for leg day anyway, might as well hit all the angles.

Different rep ranges are better suited to different movements

MJ exercises by definition utilize multiple joints and muscle groups, which can have some important implications for rep ranges. They’re really well suited to heavier loading to bring on some serious mechanical tension in the muscle fibres, and moderate rep ranges typically work well also. But what about higher reps? Do you really want to do sets of 15-30 close to failure on things like bench press or squats? A lot of times, technique breakdown, global pain/discomfort, or the ability to breathe become the limiting factors in such a scenario. You might be better off choosing an exercise with lesser technical and cardiovascular demand that also allows you to focus on the specific muscle: something like a dumbbell fly or leg extension. At the same time, those SJ exercises are typically well suited to higher rep ranges, but poorly suited for low reps. It’s probably not the best idea to test your dumbbell fly 5RM.

Stimulus to fatigue ratio

A huge part of lifting is managing fatigue levels. It’s probably a good idea to include a variety of exercises and a variety of rep ranges in your training program, and those things will bring with them a variety of levels of fatigue. To grow your lats, heavy weighted pull-ups make an awesome choice, but also come with a large fatigue cost. High rep machine pullovers can be good too, with less global fatigue, but arguably less stimulus – we’re totally speculating here, but I don’t think many people would argue that you’ll reach your peak lat size with only high rep pullovers. As a rough, anecdotal generalization (to my knowledge, there’s no research on this), MJ exercises bring more fatigue then SJ, while they may or may not bring more stimulus to the table. If you’re trying to maximize your amount of stimulus while minimizing the amount of fatigue within your entire program, we need to find the right balance of MJ and SJ exercise. There’s no one-size-fits-all answer here, but with some trial and error, you can find that balance for yourself. It should go without saying, but another consideration is for a situation like the biceps, which don’t act as a prime mover during any MJ movement. Doing only rows to try to grow your biceps is going to come with a lot more fatigue than doing curls instead. To be clear, we’re not just piling a ton of SJ work on top of a ton of MJ work – everything has to be balanced within your recovery/adaptive abilities. Think of filling up a bucket; the more MJ work you’re doing, the less room there is for SJ, and vice-versa.

Overload potential

There are multiple ways we can apply overload – weight on the bar, repetitions, sets, and proximity to failure being the most common – and while I’m a firm believer that we can overload any exercise depending on our creativity, some of these overload applications are better suited to different exercises. You can add reps to just about anything, but it’s typically easier to add load with MJ, and typically safer to get closer and closer to failure with SJ. I’m not here to argue if certain forms of overload are superior to others – I’m all about pragmatism, and if we can use them all to see improvements, then why wouldn’t we? This is also a great opportunity to flex your creativity in programming. Maybe you start your shoulder training with a heavy barbell overhead press, and then move on to some lateral raises and face pulls. You could keep the sets, rep range, and RIR targets on the OHP static and slowly add weight to the bar over the weeks, while for the lateral raises and face pulls, where it’s much more difficult to add load, you walk the reps up over the weeks until you reach a point where you can drop them back down to where you started but use a heavier load, and repeat the process.

Time efficiency

Another tradeoff to consider is time efficiency. If time is the limiting factor, MJ training is probably going to be the best bet since you can hit more muscle groups at once.

Fun

Fun is an often-overlooked aspect of programming. Adherence underlies success with all things, and if you don’t enjoy your training then you’re not very likely to adhere as the weeks, months, years, and decades pass by. Maybe you know that squats are beating you up enough that you might be able to do more total volume if you did a couple less sets and spent that time on the leg extension instead, but gee-whiz you just really love squatting and leg extensions are super boring. Maybe you’re willing to accept slightly inferior biceps growth because you just absolutely hate doing curls for some strange reason (not judging… okay I am a little bit, ya weirdo).

When it’s all said and done, both MJ and SJ exercises have their place. Here’s my rough guideline based on the literature: if a muscle acts as a prime mover and goes through a large range of motion during MJ exercise, SJ work may not be needed, but you can substitute some in and probably get a benefit either from working the muscle from a different angle or from improving the overall stimulus to fatigue ratio of the session; if the muscle never acts as a prime mover, nor goes through a large range of motion during MJ exercise, SJ work is most likely a good idea.

References

1.         Barbalho, M, Coswig, V, Raiol, R, Fisher, J, Steele, J, Bianco, A, et al. Single joint exercises do not provide benefits in performance and anthropometric changes in recreational bodybuilders. Eur J Sport Sci20: 72–79, 2020.

2.         Barbalho, M, Coswig, V, Raiol, R, Steele, J, Fisher, J, Paoli, A, et al. Effects of Adding Single Joint Exercises to a Resistance Training Programme in Trained Women. Sports6: 160, 2018.

3.         Barbalho, M, Coswig, VS, Raiol, R, Steele, J, Fisher, JP, Paoli, A, et al. Does the addition of single joint exercises to a resistance training program improve changes in performance and anthropometric measures in untrained men? Eur J Transl Myol28: 7827, 2018.

4.         Barbalho, M, Gentil, P, Raiol, R, Fisher, J, Steele, J, and Coswig, V. Influence of Adding Single-Joint Exercise to a Multijoint Resistance Training Program in Untrained Young Women: J Strength Cond Res1, 2018.

5.         Brandão, L, de Salles Painelli, V, Lasevicius, T, Silva-Batista, C, Brendon, H, Schoenfeld, BJ, et al. Varying the Order of Combinations of Single- and Multi-Joint Exercises Differentially Affects Resistance Training Adaptations. J Strength Cond Res, 2020.

6.         de França, HS, Branco, PAN, Guedes Junior, DP, Gentil, P, Steele, J, and Teixeira, CVLS. The effects of adding single-joint exercises to a multi-joint exercise resistance training program on upper body muscle strength and size in trained men. Appl Physiol Nutr Metab Physiol Appl Nutr Metab40: 822–826, 2015.

7.         Gentil, P, Soares, S, and Bottaro, M. Single vs. Multi-Joint Resistance Exercises: Effects on Muscle Strength and Hypertrophy. Asian J Sports Med6, 2015.Available from: http://asjsm.com/en/articles/21609.html

8.         Gentil, P, Soares, SRS, Pereira, MC, da Cunha, RR, Martorelli, SS, Martorelli, AS, et al. Effect of adding single-joint exercises to a multi-joint exercise resistance-training program on strength and hypertrophy in untrained subjects. Appl Physiol Nutr Metab Physiol Appl Nutr Metab38: 341–344, 2013.

9.         Mannarino, P, Matta, T, Lima, J, Simão, R, and Freitas de Salles, B. Single-Joint Exercise Results in Higher Hypertrophy of Elbow Flexors Than Multijoint Exercise. J Strength Cond Res, 2019.

10.      Ogasawara, R, Loenneke, JP, Thiebaud, RS, and Abe, T. Low-Load Bench Press Training to Fatigue Results in Muscle Hypertrophy Similar to High-Load Bench Press Training. Int J Clin Med04: 114–121, 2013.

11.      Ogasawara, R, Yasuda, T, Ishii, N, and Abe, T. Comparison of muscle hypertrophy following 6-month of continuous and periodic strength training. Eur J Appl Physiol113: 975–985, 2013.

12.      Paoli, A, Gentil, P, Moro, T, Marcolin, G, and Bianco, A. Resistance Training with Single vs. Multi-joint Exercises at Equal Total Load Volume: Effects on Body Composition, Cardiorespiratory Fitness, and Muscle Strength. Front Physiol8: 1105, 2017.

13.      R. A. Rogers, R. U. Newton, K. P. Mcevoy, E. M. Popper, B. K. Doan, and J. K. Shim. The effect of supplemental isolated weight-training exercises on upper-arm size and upper-body strength. NSCA Conf369, 2000.

14.      Schoenfeld, BJ, Contreras, B, Tiryaki-Sonmez, G, Wilson, JM, Kolber, MJ, and Peterson, MD. Regional Differences in Muscle Activation During Hamstrings Exercise: J Strength Cond Res29: 159–164, 2015.

15.      Vigotsky, A.D., Nuckols, G.L., Heathers, J., Krieger, J., Schoenfeld, B.J., and Steele, J. Improbable data patterns in the work of Barbalho et al. , 2020.Available from: https://osf.io/preprints/sportrxiv/sg3wm/